Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
University of Nebraska - Lincoln University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln
Dissertations, Theses, and Student Research from the College of Business Business, College of
2010
REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT
Adrian Chan University of Nebraska at Lincoln
Follow this and additional works at: https://digitalcommons.unl.edu/businessdiss
Part of the Business Administration, Management, and Operations Commons, Industrial and
Organizational Psychology Commons, Organizational Behavior and Theory Commons, and the Social
Psychology Commons
Chan, Adrian, "REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT" (2010). Dissertations, Theses, and Student Research from the College of Business. 8. https://digitalcommons.unl.edu/businessdiss/8
This Article is brought to you for free and open access by the Business, College of at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Dissertations, Theses, and Student Research from the College of Business by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.
REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT
by
Adrian Y.L. Chan
A DISSERTATION
Presented to the Faculty of
The Graduate College at the University of Nebraska
In Partial Fulfillment of Requirements
For the Degree of Doctor of Philosophy
Major: Interdepartmental Area of Business (Management)
Under the Supervision of Professor Fred Luthans
Lincoln, Nebraska
May, 2010
REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT
Abstract
Adrian Y.L. Chan, Ph.D
University of Nebraska, 2010
Advisor: Fred Luthans
This study proposes that being reflective or ruminative about one's leadership
experience can have differential effects on one's leadership efficacy, implicit leadership
theories and psychological capital. Specifically, through the aid of an event history
calendar, conscript military trainees of high and low military experience from a SE Asian
military organization were randomly assigned to recall and reflect or ruminate on his past
leadership experience. Results show that type of reflection interacts with level of military
leadership experience to differentially affect one's leadership efficacy, implicit leadership
theories and leadership self-awareness. Reflection triggers produced significantly higher
levels of implicit leadership theories under both low and high leadership experience
conditions. In the low leadership experience condition, reflection triggers resulted in
higher levels of leadership efficacy and leadership self-awareness but was not significant.
On the other hand, in the high leadership experience condition, the reflection triggers
performed significantly worse than the rumination triggers. Implications for the findings
are discussed.
iii
Dedication
In the course of completing this dissertation, I had to justify for time off from
work; fight for funding; navigate the trickiness of getting this dissertation done from
across the oceans; fight bureaucracy for access to data; overcome cynicism over the value
of this dissertation, first from others, and occasionally, from demons within; learn to read
without depth perception as I adapt to being afflicted with Bells Palsy; learn to walk on
crutches as I endured though having ACL reconstruction done while in the States; juggle
between conflicting role responsibilities as a researcher, academic, military officer, son,
husband, father, often not in that order or more often than not, all at once. But to stop
here would have been to stay ruminative.
I also had the privilege of experiencing the joys of fatherhood, as I watch my
children grown up as true daughters and sons of Nebraska; fret over their adjustment (no
playgrounds!) to Singapore; experience the unflagging love and devotion of my wife as
she, having gone through the dissertation process years ahead of me, has been honed by
that process to be far wiser and infinitely more patient than I could ever be being still in
the situation. Most importantly, I dedicate this dissertation to God, for whom nothing
which is accomplished by human hands means anything, and anything accomplished by
His hands reap infinitely richer rewards, some of which are appreciated now through
human eyes, while others can only be redeemed at the appropriate time when our finite
wisdom attains His holy foolishness. As such, I dedicate this dissertation to God, my wife
and my children.
iv
Acknowledgement
I thank Prof Bruce Avolio for having triggered in me this wonderful madness to
pursue a PhD at a time when I was happily settled in my military career. Most jolts are
dislocating at the start, but when carefully negotiated, they produce a rich reward. Having
worked with Bruce is one such reward.
I thank Prof Fred Luthans for chairing my dissertation committee. My style of
working is discomforting, as I tend to trek into the academic underbrush on my own for
long periods without keeping him informed. The best thing about Fred is his wonderful
sense of pragmatism that keeps me anchored, and his sense of completion that enables me
to pull through and get this done at the moment where it mattered. Without Fred, I‘d still
be floundering in the underbrush!
I thank Profs Mary Uhl-Bien and Peter Harms. Between them, they weighed in
with solid comments to challenge me to greater professional standards. They represent
for me the exacting standards that all aspiring scholars should uphold. From them I have
learnt that to become the respected members of the academic community that they are,
one must sweat the small stuff, must not compromise on the right stuff, and must be made
of sterner stuff than all that life can throw at you. Thank you Mary, thank you Peter!
Lastly, I thank Prof Cal Garbin over at the Psychology Dept. He started me on the
road to loving research methods and stats that have stood me well, whether it is for this
dissertation or for my future endeavors. I hope he continues to spread the love for a long,
long time!
v
Table of Content
Abstract .......................................................................................................................... ii
Dedication .................................................................................................................... iii
Acknowledgement ........................................................................................................ iv
Table of Content ............................................................................................................ v
Chapter 1: Introduction to the dissertation................................................................... 13
Rationale of current research ................................................................................... 13
Research Aims ......................................................................................................... 15
Summary of key research questions ........................................................................ 17
Chapter 2: A review of the literature ........................................................................... 19
The problematic nature of leadership experience .................................................... 20
Levels of leadership experience ............................................................................... 23
Psychological Capital and Leadership Experience .................................................. 26
Leadership Efficacy and Leadership Experience ..................................................... 28
Leadership Self-Awareness and Leadership Experience ......................................... 29
Leader Portrait. .................................................................................................... 30
Leader Purpose..................................................................................................... 31
Leadership Paradigm. .......................................................................................... 32
Implicit Leadership Theories and Leadership Experience ....................................... 33
Role of self-reflection in transforming leadership experience ................................. 35
Reflection, experience, experiential learning and learning styles ............................ 37
Reflection as a trait and as a learning activity ......................................................... 41
Adaptive versus maladaptive forms of reflection .................................................... 42
Ways of invoking reflection..................................................................................... 45
Experience, reflection and leadership efficacy ........................................................ 47
vi
Reflection and Leadership Self-Awareness ............................................................. 49
Reflection and Implicit Theories of Leadership ...................................................... 51
Summary and integration of literature ..................................................................... 53
Chapter 3: Study Design and Methodology ................................................................. 56
Research Design....................................................................................................... 56
Sampling Plan and Power Analysis. .................................................................... 56
Participants and Training Context ........................................................................... 57
Pilot Studies ............................................................................................................. 57
Pilot Study 1 ......................................................................................................... 58
Scale-creation: Quality of Reflection Outcome Scale ............................................. 61
Pilot Study 2. ........................................................................................................ 62
Missing data ............................................................................................................. 62
Exploratory Factor Analyses................................................................................ 63
Confirmatory Factory Analyses. .......................................................................... 66
Measures .................................................................................................................. 68
New Measure: Quality of Reflection Outcomes. ................................................. 68
Generalized Leadership Efficacy. ........................................................................ 69
Psychological Capital (Psycap)............................................................................ 70
Trait Reflectiveness. ............................................................................................ 70
General Self Efficacy. .......................................................................................... 73
Sentence Completion Exercises: Implicit Theories of Leadership and Leadership
Development. ....................................................................................................... 74
Procedure and Manipulation .................................................................................... 76
Reflection Manipulation. ..................................................................................... 77
Rumination Manipulation. ................................................................................... 77
vii
Control. ................................................................................................................ 77
Manipulation Checks ............................................................................................... 78
Reflection/Rumination invokes emotional processing ........................................ 78
Rumination intervention is valid .......................................................................... 80
Reflection intervention is valid ............................................................................ 81
Establishing initial equivalence. .......................................................................... 82
Chapter 4: Results ........................................................................................................ 84
Data Screening ......................................................................................................... 84
Outlier Analysis ....................................................................................................... 84
Normality ................................................................................................................. 86
Initial equivalence testing ........................................................................................ 87
Hypotheses Testing .................................................................................................. 87
Hypotheses 1a-1d ..................................................................................................... 88
Hypotheses 2a-c ....................................................................................................... 89
Interpreting the MANCOVA results. ................................................................... 90
Post-hoc Explorations & Hypotheses 3-5 ................................................................ 91
Low leadership experience. ................................................................................. 92
High leadership experience. ................................................................................. 93
Predictors of GLE at Time 2 for high leadership experience .............................. 93
Chapter 5: Discussion .................................................................................................. 97
Review of Purpose of Dissertation .......................................................................... 97
Overview of Discussion of Results .......................................................................... 97
Effect of Leadership Experience: Discussion of Hypotheses 1a-d .......................... 98
Lack of effect on Psycap ...................................................................................... 98
Lack of effect on leadership self awareness. ....................................................... 99
viii
Interaction of Leadership Experience and Type of Reflection Triggers: Discussion
of Hypotheses 2 to 5 ................................................................................................ 99
Implicit leadership theories. ............................................................................... 100
Leadership efficacy. ........................................................................................... 100
Leadership self-awareness. ................................................................................ 102
Post-Hoc Analyses: Predictors of leadership self-awareness. ........................... 103
Potential Limitations .............................................................................................. 104
Climate. .............................................................................................................. 104
Effectiveness of the reflection-rumination manipulations. ................................ 105
Participants‘ motivation. .................................................................................... 105
Uneven Sample Distribution and Power. ........................................................... 106
Measurement Limitation. ................................................................................... 106
Method Limitations. ........................................................................................... 106
Experimenter Bias. ............................................................................................. 107
Sample................................................................................................................ 107
Training Context. ............................................................................................... 107
Recommendations and Implications for Future Research ..................................... 108
Developmental climate. ..................................................................................... 108
Competing demands........................................................................................... 110
Reflective practices and habits........................................................................... 111
Developmental readiness and motivation to lead. ............................................. 111
Theoretical and Practical Contributions................................................................. 111
Reflection is not enough. ................................................................................... 112
Distinguishing skill development and identity development. ............................ 112
Motivation and identity. ..................................................................................... 113
ix
Conclusion ............................................................................................................. 113
Figures........................................................................................................................ 114
Figure 1: Hypothesized Model............................................................................... 114
Figure 2: 3-stage model linking adaptive reflection to experience, development and
performance ........................................................................................................... 115
Figure 3: Effect of Time 1 Predictors on Quality of Reflection at Time 2 ............ 116
Figure 4: Interaction of Military Leadership Experience with Type of Reflection on
Leadership Self-Awareness ................................................................................... 118
Figure 5: Interaction of Military Leadership Experience with Type of Reflection on
Leadership Efficacy ............................................................................................... 119
Figure 6: Interaction of Leadership Experience with Type of Reflection on Quality
of Implicit Leadership Theories ............................................................................. 120
Figure 7: Predictors of Generalized Leadership Efficacy at Time 2 ..................... 121
Tables ......................................................................................................................... 123
Table 1: Events that provide highest proportions of developmental triggers ........ 123
Table 2: Coding Frame for evaluating reflections ................................................. 124
Table 3: EFA results .............................................................................................. 125
Table 4: EFA item loadings for Quality of Reflection Scale in Pilot Sample ....... 126
Table 5: EFA Item Loadings for Trait Reflectiveness Scale in Pilot Sample ....... 128
Table 6: Descriptive Statistics for Manipulation Check Item 1 ............................. 129
Table 7: Descriptive Statistics for Manipulation Check Item 2 ............................. 130
Table 8: Descriptive Statistics for Manipulation Check Item 3 ............................. 131
Table 9: Descriptive Statistics for Variables in Basic Military Course Study Sample
(Low Leadership Experience) ................................................................................ 132
x
Table 10: Descriptive Statistics for Variables in Leadership Course Study Sample
(High Leadership Experience) ............................................................................... 133
Table 11: Skewness and Kurtosis Statistics for Variables in Basic Military Course
Study Sample (Low Leadership Experience) ........................................................ 135
Table 12: Skewness and Kurtosis Statistics for Variables in Leadership Course
Study Sample (High Leadership Experience) ........................................................ 136
Table 13: Summary of Hypotheses ........................................................................ 138
Appendices ................................................................................................................. 140
Appendix A: Sequence of Activities at Time 1 & 2 .............................................. 140
Appendix B: Measures ........................................................................................... 141
Measure 1: Quality of Reflection ....................................................................... 141
Measure 2: Generalized Leadership Efficacy .................................................... 143
Measure 3: Psycap Scale (4 Components) ......................................................... 145
Measure 4: Trait Reflectiveness Scale ............................................................... 147
Measure 5: New General Self-Efficacy Scale (Chen, Gully & Eden, 2001) ..... 149
Manipulation Check items ................................................................................. 150
Outcome Measure 1: Sentence Completion Exercise (Implicit Leadership
Theory) ............................................................................................................... 151
Outcome Measure 2: 5-Item Goal Setting Exercise .......................................... 152
Appendix C: Prompts to prime Reflection And Rumination ................................. 153
References .................................................................................................................. 154
11
REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT
―The cloak of the past is cut from patches of feeling, and sewn with rebus
threads. Most of the time, the best we can do is wrap it around ourselves
for comfort or drag it behind us as we struggle to go on. But everything
has its cause and its meaning. Every life, every love, every action and
feeling and thought has its reason and significance; its beginning, and the
part it plays in the end. Sometimes we do see. Sometimes, we see the past
so clearly, and read the legend of its parts with such acuity, that every
stitch of time reveals its purpose, and a kind of message is enfolded in it.
Nothing in any life, no matter how well or poorly lived, is wiser than
failure or clearer than sorrow. And in the tiny, precious wisdom that they
give to us, even those dread and hated enemies, suffering and failure, have
their reason and their right to be.‖
- ‗Shantaram‘, pg 871-872 (Gregory David Roberts)
12
CHAPTER 1: INTRODUCTION TO THE DISSERTATION
13
REFLECTING ON EXPERIENCE FOR LEADERSHIP DEVELOPMENT
Chapter 1: Introduction to the dissertation
The purpose of this dissertation is to explore how adaptive versus ruminative
ways of reflecting on past experience can influence either better or poorer ways of
conceptualizing one‘s leadership, of being more or less well-developed in one‘s naïve
theories about leadership, becoming more or less efficacious regarding one‘s ability to
lead, and higher or lesser levels of psychological capital.
Rationale of current research
Why should one be concerned about how leaders reflect on past experience?
There are several reasons why this research is of value to the leadership development
domain.
Firstly, operationalizing leadership experience merely in terms of traditional
demographic variables such as tenure and supervisory experience is insufficient. Rather,
this dissertation proposes that the value in having these experiences lies in how such
experiences are influenced by reflection for better leadership outcomes.
Secondly, while it is intuitive that reflection is important for leadership
development, and indeed features in many leadership development efforts, exactly how
this reflection is done is left very much to the imagination of the LD practitioner or
leader. Does it make a difference if the reflection is ruminative, focused on self, or
focused on problem-solving?
14
In line with the research aims of the present dissertation, there are several related
issues that will also be explored. Firstly, what constitutes relevant experience, especially
when selecting for leaders who are expected to lead in novel situations with ill-defined
problem parameters, e.g. in creative environments or situations with conflicts of interests
that are not clear? What experience has developmental potential? In a training context,
are the experiences designed by trainers for development interpreted and processed in a
similar fashion by the intended trainees?
The second issue concerns the processes by which leaders can reframe past
experience for leadership development. The issue of relevant experience is turned on its
head – what if experience is controlled while allowing how the experience is processed to
vary? Can one generate new leadership insight simply by varying the way these past
experiences have been processed?
Obviously, not all the research questions raised above can be adequately
addressed in any single study. Nonetheless, these questions provide a peek into the line of
inquiry in the research agenda of which this dissertation is a part of. If indeed the way
leaders reflect on past experience can make them better leaders, then this in itself is a
worthy outcome.
Another potential contribution of this dissertation is to help achieve a better
understanding and application of reflection for leadership development. Reflection
represents a cost-efficient means of leadership development that promises high return on
development (ROD) for organizations. Through reflection, no expensive training
investment is needed as the ‗raw materials‘ for development is already available and
present in the leaders - all it takes is a better way of processing these experiences.
15
Finally, by reflecting on existing experiences, this dissertation promotes
leadership development to be self-sustaining, since these experiences are available to
leaders to process at their own pace. In addition, leader selection may be improved with
the new insight into how, why and under what conditions leadership experience is useful.
Research Aims
------------------------------------------------------------
Insert Figure 1 Here
------------------------------------------------------------
Refer to Figure 1. This dissertation seeks to demonstrate empirical linkages
between leadership experience, implicit leadership theories, leadership efficacy, self-
awareness and psychological capital. Firstly, the literature is reviewed to examine the
impact of leadership experience on leadership development and performance, broadly
operationalized as encapsulating psychological capital, leadership efficacy, leader self-
awareness and leader implicit theories of leadership. An observation derived from this
review is that on its own, leadership experience does not adequately explain variations in
leadership development and performance. An approach taken by some scholars to address
this limitation was to identify situations and moderating variables which help to examine
how leadership experience predicts leadership development and performance more
adequately. This approach continues to assume a direct causal relationship between type
and nature of past leadership experience and future leadership development and
performance.
The approach taken by this dissertation is to propose an alternative model that
focuses on processes that transform leadership experience. This presumes the mediational
16
role of reflection between past leadership experience and future leadership development
and performance, moderated by different types of reflection.
In this process model, leadership experience provides the raw materials; reflection
helps transform the experience into insights. High quality reflections help a leader feel
more self-aware and self-confident as a leader. The insights themselves may contribute to
greater psychological capacities for future leadership performance.
In addition, this dissertation distinguishes between two types of reflection –
adaptive and maladaptive. The terms ‗adaptive‘ and ‗maladaptive‘ are descriptors derived
from the clinical psychology literature (see review section later). The process of adaptive
reflection involves a strong focus on the issues, characterized by problem-solving,
experiential mindfulness and openness. People engaging in such reflections are
hypothesized to be more adaptive in overcoming adversity. The process of maladaptive
reflection is more self-focused and ruminative, typically triggered by perceived threats to
oneself rather than by an intrinsic curiosity for discovery. Such reflections tend to dwell
on the effects or symptoms of a problem on oneself rather than on finding solutions to the
problem. People engaging in such reflections are associated in literature with more
maladaptive reactions to adversity. In this dissertation, I will refer to adaptive reflection
as reflection, while maladaptive reflection is referred to as rumination. Both types of
reflection will be referred to as reflective processes.
This dissertation tests the hypotheses that different types of reflection will
moderate the effect of leadership experience on leadership development and
performance. Proximal outcomes such as leadership efficacy, activations of one‘s implicit
leadership theory and leader self-awareness (operationalized by quality of leadership
17
reflection outcomes) are hypothesized to respond differently to different types of
reflections while positive psychological states, referred to hereafter as psychological
capital, due to it being a positive resource that is built up through repetitions, may be less
susceptible to one-off interventions such as those provided in this dissertation.
Finally, while not explicitly manipulated in this dissertation, exploratory analyses
will be conducted to examine the role that initial leadership efficacy has on how leaders
reflect on experience. It is postulated that low initial leadership efficacy may be more
susceptible to maladaptive forms of reflection than high initial leadership efficacy.
Hence, within the group in which maladaptive reflection was introduced, participants
whose initial leadership efficacy was low may experience in greater measure the signs
that the maladaptive trigger has worked.
Summary of key research questions
The key research questions under investigation in this dissertation are:
1. What is the impact of adaptive and maladaptive reflection on one‘s leadership
efficacy, one‘s implicit leadership theories and leader self-awareness that get
activated (operationalized as quality of leadership reflection outcomes), and
one‘s psychological capital?
2. Does higher leadership experience affect the above outcomes differently?
18
CHAPTER 2: A REVIEW OF THE LITERATURE
19
Chapter 2: A review of the literature
“By three methods we may learn wisdom: First, by reflection, which is noblest; second,
by imitation, which is easiest; and third by experience, which is the bitterest.”
- Quote attributed to Confucius
There are some who say that the practice of leadership is an art (Conger, 1989; De
Pree, 2004; Fairhurst & Sarr, 1996). To the extent that art is about the effective
application of a body of knowledge, then leadership is in that sense an artful application.
Yet while it may be possible to duplicate masterpieces using technology,
leadership episodes are often unique, making it difficult, if not impossible, to replicate in
its entirety. More often than not, particularly in high reliability organizations, leaders do
not have the luxury of repeating a leadership episode, much less repeat a mistake, in
order to (re)learn from it (LaPorte & Consolini, 1991). It is also often the case that
leaders are called to perform in difficult or novel situations that require them to act wisely
(Robert J. Sternberg, 2003), or to improvise and adapt from experience (Michael D.
Mumford & Connelly, 1991). In fact, leadership itself is an adaptive process (DiTomaso
& Hooijberg, 1996; Heifetz, 1994).
Evidently, leaders who act wisely will be highly valued. Since wisdom is one of
the few personal strengths that increase with age and experience rather than decrease
(Baltes & Smith, 1990; Denny, Dew, & Kroupa, 1995), this may explain why when
selecting for leaders, recruiters invariably look for experience as a criterion for selection
20
(B. K. Brown & Campion, 1994). However, surely not all leaders, especially the more
junior leaders, come ‗packaged‘ with the ‗right‘ experience – or do they?
The problematic nature of leadership experience
What is leadership experience? How does it differ from life experience? Why is it
important and when is it important? There is intuitive appeal to the notion that leadership
experience should play an important role in predicting future leadership development and
performance. From a human resource perspective, leadership experience plays a part in
job interview success (Tay, Ang, & Dyne, 2006), Yet, as will be seen in the review
below, it is not always clear how and when leadership experience is important, or even if
the way it has been operationalized actually captures the full breadth of the construct.
As early as 1970, Fred Fiedler concluded from several studies that there was only
a low correlation of .12 between supervisory experience and leadership performance (F.
E. Fiedler, 1970). Two explanations were put forth by Fiedler to account for this counter-
intuitive low correlation. One was methodological – that the low correlation could be due
to the narrow construct definition of experience operationalized solely in terms of
supervisory tenure in the studies. In other words, not all forms of experience that are
useful for predicting performance may have been captured. When the ‗correct‘ type of
leadership experience is captured, one would expect prediction to improve. For example,
when considering the prior experience of baseball coaches, which is experience that is
most similar to current job demands faced by the baseball coaches, team success is more
reliably predicted (Cannella & Rowe, 1995).
The other explanation offered by Fiedler was theoretical – that there were
contingency factors at play which were not taken into account (Fred E. Fiedler, 1972).
21
For example, when the job situation contains a succession situation, then leader
experience is no longer predictive of performance (Pfeffer & Davis-Blake, 1986).
When multiple contingencies are introduced, the situation becomes even more
complicated. For example, in a series of field studies that Fiedler conducted, the presence
of stress and leader intellectual abilities combined to produce counter-intuitive findings to
the role of experience in predicting performance. He found that under low stress, leader
experience was negatively correlated with performance, while under high stress, it was
positively correlated. Conversely, leader intellectual abilities correlated with performance
in the opposite direction under similar stress conditions (F. E. Fiedler, McGuire, &
Richardson, 1989).
An interesting twist to the concept of leader experience was provided by Fiedler.
Fiedler in his later works where he conceptualized experience as a form of over-learned
behavior that takes over higher cognitive processing when cognitive resources are scarce,
for example such as in situations of high stress in which limited cognitive resources are
available for performance (F. E. Fiedler, 1995, 2002). One way to encourage the
development of such over-learned behavior is through training, particularly training that
utilizes behavioral modeling which focuses on learning and repetition of desired
behaviors (Taylor, Russ-Eft, & Chan, 2005). However, even if the leader possesses the
requisite attributes such as experience, intellect and training, other contingency factors
such as group climate and communication styles may still act to prevent the experienced
leader from performing well (Murphy, Blyth, & Fiedler, 1992).
At the other end of the equation, the nature of the task is also an important factor.
When the task criteria are similar to those encountered in one‘s past experience, then
22
leadership experience becomes predictive of success (Gabarro, 1987; Locke &
Kirkpatrick, 1991). Conversely, when the task criteria are unstructured and ambiguous,
leadership competencies, particularly problem-solving skills, rather than the actual
experience itself may be more important (Cannella & Rowe, 1995; M. D. Mumford,
Zaccaro, Harding, Jacobs, & Fleishman, 2000). Yet this does not suggest a mutually
exclusive set of attributes for leaders – both experience and competencies are important,
and are probably mutually reinforcing. Developmentally, experience plays a role in
honing these competencies (Copeman, 1971; Lindzey, Holmes, & McCall, 1987). At the
same time, it takes specialized competencies such as wisdom to properly utilize
experience for novel situations (R.J. Sternberg, 2004).
Hence, the role of leadership experience on performance is contingent on the type
of performance needed, and the operating conditions under which the leader is required to
function. An implicit assumption that was not challenged was that there was a direct
relationship between leadership experience and subsequent performance. As will be seen
later, this dissertation proposes an alternative approach of identifying variables that
directly and indirectly influence the experience-performance equation.
In any case, the complex interplay of leadership experience and contingent factors
was a topic of continuing concern to Fiedler. At one point, it even caused him to remark
that ―no consistent relationship between experience or job tenure and leadership
performance‖ can be found (Fred Edward Fiedler & Garcia, 1987, p.41).
Given this state of affairs from the contingency perspective, some scholars have
turned to a more process-focused explanation for the utility of leadership experience. For
example, when previous leadership experience was evaluated for its relevance to current
23
requirements, it positively predicted performance (Bettin & Kennedy, 1990). When
experience was conceptualized in terms of tacit knowledge drawn from learning from
daily experiences, it was shown to be predictive of leadership performance in procedural
tasks (Hedlund et al., 2003). Underpinning these two studies are meaning-making
processes that enable leadership experience to undergo some form of transformation
(Mezirow, 2000a). From a learning perspective, experience becomes relevant and useful
when lessons are learnt from it and it becomes part of the leader‘s protocol of actions and
understanding of leadership. For lessons to be learnt, experience needs to undergo the
dual process of ―prehension‖ (or ‗grasping‘) and transformation (D. A. Kolb, 1984).
Experience then becomes raw material upon which one can reflect (Dewey, 1933; Schön,
1983). The role of reflection in experience will be covered in more detail in later sections
of the dissertation.
Levels of leadership experience
As a prelude to addressing the above issues, there is a need to first examine the
levels by which leadership experience can be conceptualized. At the group level, the
leadership experience can be a shared phenomenon with group members. In such
situations, leaders can demonstrate learning by virtue of their membership in a larger
collective (Edmondson, 1999). Again at the group level, the leadership experience may
also be asymmetric – while a common experience is shared, the outputs derived may
differ between leaders and other members of the group. One example of such asymmetry
can be found in leaders who lead organizations that are themselves exhibiting learning
behaviors (Argyris, 1999). Argyris argues that the outputs differ depending on the type of
learning that occurs (e.g. single versus double looped). At the individual level, leadership
24
experiences may be regarded as an ipsative process. For example, the experiential
perspective to adult learning assumes a vital interaction between actor and the
environment (Malinen, 2000), and that each interaction and impact is unique to the actor.
Finally, common to all levels of conceptualization is an intra-personal level perspective.
It is at the intra-personal level that the transformation of experience to knowledge
occurs. This dissertation focuses on the nature of these transformative processes for
making life experience useful and relevant for leadership development and performance.
Through these transformations, knowledge acquired in one aspect of a leader‘s
experience can be made useful for a leadership-related situation.
Using the analogy of the firm, a leader can be described as a ‗one-person firm‘
that can benefit from the transfer of knowledge that occurs between one part of an entity
and another part of the same entity that needs it (Argote, 2005; Reagans & McEvily,
2003). Similarly, a leader‘s past experience, regardless of whether it is directly related to
leadership, can provide valuable knowledge that may be transferred to other aspects of
the leader‘s current experience.
Collective learning research adopts a definition of learning that includes asking
questions, challenging assumptions, seeking different perspectives, evaluating
alternatives, and reflecting on past actions (e.g. Edmondson, 2002; Gibson & Vermeulen,
2003). These aspects of learning are relevant to the group as well as to the individual, and
run consistent with a process-focus approach to explaining why experience is useful for
performance. A particular focus of this dissertation is on the type of learning behavior
known as critical reflection (Dewey, 1933), which will be reviewed in later sections.
25
The transferability of knowledge in leaders has important consequences for
leadership development and performance. Leaders need to be adaptive and continually
learning so as to perform well. In fact, it is the uniqueness and adaptive nature of
leadership that has led some scholars to contend that leadership occurs only in the face of
adaptive challenges or problems (Heifetz, 1994). Yet surprisingly, in spite of its
importance, our understanding of how leaders learn and adapt is still largely theoretical
(Day, 2000; DiTomaso & Hooijberg, 1996; R. G. Lord & Hall, 2005). For an exception
regarding empirical research, see Hirst, Mann, Bain, Pirola-Merlo and Richver (2004).
Likewise, the adaptive nature of leadership is not high on one‘s schema of a leader across
many cultures – for example, it is not listed as part of culturally implicit schemas of
leadership attributes (House, Hanges, Javidan, Dorfman, & Gupta, 2004).
Given the introductory discussion on leadership experience thus far, one would
expect that leaders with greater leadership experience (in temporal terms) should be
better developed leaders and perform better than leaders with less experience. The
question is: better developed how and performs better in what ways? This dissertation
operationalized such development and performance to be associated with leadership
efficacy, psychological capital, leader self-awareness, and implicit theories of leadership.
Reasons for such an operationalization will be discussed in later sections.
Thus far, an implicit assumption in this discussion is that all forms of leadership
experiences be they positive ones or negative ones, are in some ways ‗good‘ experiences.
It is obvious how positive experiences directly contribute to future performance. For
example, positive experiences give rise to higher efficacy beliefs that in turn predict
future performance (Bandura, 1997). (More on efficacy beliefs will be reviewed in a later
26
section of this chapter). However, it is less clear how negative experiences can be
processed in positive ways so that they too become ‗good‘ lessons for the leader to have
‗under his/her belt‘, metaphorically speaking. The next section of this review explores
how reflection is one such mechanism for processing experience.
Psychological Capital and Leadership Experience
Leaders are also followers, or have been followers at some point in their
development. Good followers demonstrate good citizenship behaviors, and good
followership is a precursor to leadership (R.E. Kelley, 1992; R. E. Kelley, 1998).
Recently, a new approach towards positively managing human resources has been
championed by Luthans and colleagues (Luthans & Youssef, 2007; Luthans, Youssef, &
Avolio, 2007). Termed as positive organizational behavior or POB, Luthans and
colleagues defined it as ―… the study and application of positively oriented human
resource strengths and psychological capacities that can be measured, developed and
effectively managed for performance improvement.‖ (Luthans, 2002, pg. 59).
A key feature of this approach is the emphasis on psychological capacities that are
measurable and developable, as well as manageable and having demonstrable linkages to
performance. The former two attributes are developmental, while the latter two are
slanted towards performance. This orientation aligns neatly with leadership development
for performance.
What constitute these psychological capacities? Luthans and colleagues
articulated inclusion criteria for these psychological capacities that included 1) evidence
of a strong theoretical and research foundation, 2) validity 3) relatively unique to the field
of organizational behavior, 4) state-like and amenable to development and 5) positively
27
impacts work performance and satisfaction(Luthans, 2002; Luthans, Avolio, Avey, &
Norman, 2007). The final set of capacities that met these inclusion criteria were hope,
resilience, optimism, and self-efficacy which have been shown empirically to aggregate
to a higher-order factor that Luthans and colleagues have named psychological capital or
Psycap (Luthans, Avolio et al., 2007; Luthans & Youssef, 2007).
Psycap refers to ―…an individual‘s positive psychological state of development
and is characterized by 1) having confidence (self-efficacy) to take on and put in the
necessary effort to succeed at challenging tasks; 2) making a positive attribution
(optimism) about succeeding now and in the future; 3) persevering towards goals and,
when necessary, redirecting paths to goals (hope) in order to succeed; and 4) when beset
by problems and adversity, sustaining and bouncing back and even beyond (resilience) to
attain success‖ (Luthans, Youssef et al., 2007, pg. 3).
From the above definition, one can infer that experience is an essential ingredient
for Psycap to increase. Efficacy beliefs are strengthened when enactive mastery is
achieved in one‘s experiences (see later section on efficacy). Attributions on success in
current experience can benefit from previous episodes of successful attributions.
Likewise, hope and resilience are enhanced with greater experience derived from having
previously navigated similar setbacks or obstacles. Because of the developmental
orientation of Psycap, it makes sense to advance the following hypothesis proposed in the
previous section:
H1a: The level of Psycap is positively correlated with prior leadership
experience.
28
Leadership Efficacy and Leadership Experience
What are efficacy beliefs? Efficacy beliefs are important determinants of behavior
– they affect choice of action, motivation to enact those behaviors and persistence of
effort in the face of adversity (Bandura, 1997). Efficacy beliefs have dimensions of
magnitude, strength, and generality (Bandura, 1997). Magnitude reflects the level of task
difficulty, while strength reflects the conviction that one can perform at that level of task
difficulty. Generality reflects the co-variation of efficacy beliefs across related domains.
Efficacy beliefs are evaluative in nature, consisting of one‘s assessments of one‘s
own abilities to successfully enact the required behaviors (Bandura, 2001). According to
Bandura (1997), efficacy beliefs are initially domain-specific and over time become
generalized across domains. For example, an athlete who has been undergoing training
for short distance running soon develops efficacy for sprinting as he undergoes success in
his training (via enactive mastery). This efficacy is initially specific to the domain of
sprinting. As confidence in one‘s athletic prowess grows, this efficacy in sprinting may
over time ‗spill over‘ to performance in track in general (i.e. ‗I am good in track.‘), or
even to the more general domain of athletic ability (i.e. ‗I am good in all sports that
require speed.‘).
Given this understanding of efficacy beliefs, leadership efficacy is defined as
one‘s beliefs in one‘s abilities to perform as a leader. It is an efficacy belief specific to the
domain of leadership. According to Bandura, domain-specific efficacy beliefs such as
leadership efficacy can be increased best by enactive mastery, followed by vicarious
learning through role models, social persuasion and physiological arousal, of which the
29
most effective is enactive mastery followed by vicarious learning (Bandura, 2000;
Stajkovic & Luthans, 1998).
Based on the above understanding of how efficacy beliefs are developed, one can
surmise that the experience of having performed well in a leadership situation will lead to
an increase in one‘s leadership efficacy. This is because performing well in a situation is
a powerful form of enactive mastery of the required leadership behaviors. Likewise,
recalling past enactments of these successful behaviors is a powerful source of vicarious
learning, due to opportunities for reviewing and learning new insights that may contribute
to better performance.
On the other hand, when one performs poorly in a leadership situation, there is
little enactive mastery afforded, which should lead to poorer efficacy beliefs concerning
leadership. When one is asked to recall such poor performances, the mental re-enactment
of these performances should similarly confirm one‘s previous beliefs concerning one‘s
leadership efficacy beliefs. In Bandura‘s terminology, no re-calibration of efficacy beliefs
occur as recall only serves to confirm these beliefs (Bandura, 1997).
H1b: The level of leadership efficacy is positively correlated with high leadership
experience.
Leadership Self-Awareness and Leadership Experience
What is leadership self-awareness? An important component of self-awareness in
a leader is being cognizant of one‘s identity as a distinct entity from others. Just as the
self-aware person is conscious of him or herself as a unique entity and builds a self-
concept over a lifetime, the self-aware leader must too be conscious of his or her identity
as a leader and be active in developing it over time. Therefore, the content and structure
30
of one‘s self-conceptions of leadership and being a leader, which has been referred to by
some leadership scholars as the ―leader self construct‖ (Hannah, Woolfolk, & Lord,
2009), has implications for the proposed definition of leader self-awareness. This leader
self construct is built from three domains which serve as the focus for what leaders
become aware of, that is, their leader portrait, purpose, and paradigm.
Leader Portrait. The first component of a leader‘s self construct is the leader
portrait. This domain is made up the leader‘s perceptions of his/her own characteristics,
qualities, and behaviors related to enacting leadership, which is composed of leadership
attributes manifested through the leader‘s behaviors. The leader portrait is basically one‘s
image of oneself as a leader, based on self construal as well as feedback received from
other people over time. This aspect of the leader self construct answers questions such as
―who am I as a leader‖ and ―what are my strengths as a leader based on self observation
and other‘s opinions‖.
Roberts and colleagues (Roberts, Dutton, Spreitzer, Heaphy, & Quinn, 2005)
introduced the idea of a person‘s reflected best self portrait that is relevant to this review.
Extending on their work, I expand the term portrait to include not only the best views, but
also information about when one is not at one‘s best, and to include self-opinions as well
as those opinions reflected from others. To be most effective in certain styles of
leadership, leaders need to understand their areas of strength and weakness as perceived
by others and from their own self-assessments (Sosik, 2001). Inherent biases exist in self-
evaluation (Ashford & Tsui, 1991) yet other‘s people opinions are not necessarily more
accurate. So a full understanding of one‘s portrait as a leader includes awareness of the
31
leader‘s self from multiple points of view, which with more leadership experience, is
more likely to be achieved.
Besides views of the self whether at one‘s best or otherwise, the leader portrait
also contains goals for the leader self, such as future ideal, possible, hoped for, or even
feared qualities that a leader may be trying to develop or avoid in him/herself (Higgins,
1987; Markus & Nurius, 1986). As leaders participate in more varied leadership episodes,
they observe themselves and other people in leader and follower roles, and from these
experiences create more self-goals for their leader development.
Leader Purpose. The second domain in the leader self-construct is a leader‘s
purpose. Each leader has motives that guide their leadership behavior, in addition to their
leader portrait and self-image just discussed. Depending on what a leader believes the
purpose of leadership is, the purpose of their own leadership affects their decisions and
behaviors as a leader. This content may overlap with a leader‘s general purpose in life
(Sosik & Dworakivsky, 1998).
A leader‘s general purpose is to create and bring to life a vision of the future. A
vision is an image of the future that helps articulate the values, purposes, and identity for
followers (Boal & Bryson, 1988). The quality of a leader‘s articulation of a desired future
state affects follower attitudes and ultimately their performance (Kirkpatrick & Locke,
1996). Hence, to be impactful, a leader needs to have high awareness of the content
appeal and degree of alignment between his/her own vision vis-à-vis those of his/her
followers. When there is alignment between leaders and followers in their values and
shared identities, the leader-follower influence mechanisms are reinforced(R.G. Lord &
Brown, 2001), particularly for certain leadership styles such as transformational
32
leadership (Jung & Avolio, 2000), which depicts this process in terms of both personal
and social identification. Naturally, this is a process which takes time and with greater
leadership experience, should improve over time.
The leader‘s purpose would contain his/her broader vision and goals that describe
the future the leader intends to create. Once the purpose is understood, leaders can
communicate this purpose and vision to their constituents, and take actions to manifest it
into reality. Again, this requires multiple interactions and long-term relationship building
over time.
Leadership Paradigm. The third component of the leader self construct is the
leadership paradigm used by the leader. Paradigm refers to the operating assumptions
regarding leadership particularly regarding the relational aspects or social roles of
leadership. Besides personal qualities contained in the leader portrait, and motives
contained in the leader purpose, the leadership paradigm contains information regarding
the kinds of questions such as, ―what assumptions do I have about how effective
leadership operates and how quality relationships are built between leaders and
followers‖ and ―what principles currently underlie my actions as leader‖.
A leader‘s conceptions about the self and their leadership paradigm are embedded
in a broader underlying set of assumptions often referred to as a ―worldview‖.
Worldviews have been described as frames of reference, paradigms, or lenses (for
review, see Koltko-Rivera, 2004), yet all terms refer to a person‘s core beliefs, values,
and assumptions that filter and shape one‘s personal meaning-making process regarding
the underlying nature of people and their social and physical reality (Dweck, Chiu, &
33
Hong, 1995; Kelly, 1955). It is reasonable to presume that with more experience, such
core beliefs and values should be better developed.
Maurer, Weiss, & Barbeite (2003) explain that worldviews are made up of
implicit assumptions that ―create a framework within which the person interprets the
world‖ (p. 108) and thus are closely linked to a leadership paradigm, and may even
contain the same assumptions in some cases. For example, Dweck and colleagues‘
research suggests that leaders who believe that personality is fixed (which they called an
entity perspective) tend to interpret people‘s behaviors in terms of traits and dispositions,
whereas leaders who view personality as developable (an incremental perspective) tend
to take situational factors into account when explaining people‘s behavior (Dweck, Hong,
& Chiu, 1993). Hence, leaders need to be ―self‖ aware about the paradigm in use to
ensure that they are appropriate for the situation at hand in terms of how they guide a
leader‘s interpretation and actions. They can then consider alternative paradigms and
potentially choose and learn to operate under different assumptions, and in this way
engage in greater self-regulation depending on the situation (Sosik, Potosky, & Jung,
2002).
Given the above review, the following hypothesis is advanced:
H1c: The level of leadership self-awareness is positively correlated with high
leadership experience.
Implicit Leadership Theories and Leadership Experience
Implicit theories of leadership are lay theories, naïve theories or mental models
that people have of leadership (Foti & Luch, 1992; R. G. Lord, Foti, & De Vader, 1984;
R. G. Lord, Foti, & Phillips, 1982). These are essentially ideas that people possess of how
34
leaders ought to be like and behave (Offerman, Kennedy Jr., & Wirtz, 1994), and may
even be extended to one‘s understanding of how the leadership process ought to be
played out.
There is some support for the role that experience plays in shaping one‘s implicit
leadership theories. Firstly, early experience may play a fundamental role in the shaping
of these implicit theories. For example, Keller (1999) identified that parental influences
may be linked to individual differences in implicit leadership theories. Hunt, Boal and
Sorenson (1990) found that early childhood experience partially explained how
leadership prototypes are categorized.
In addition, not only does experience shape the nature of these implicit theories,
the implicit theories themselves may subsequently shape the perception and sense-
making of new experience (Tiffany Keller, 2003). Evidence suggests that the implicit
theories of leadership may 1) themselves be a source of manifested leader behavior
(Wofford & Goodwin, 1994; Wofford, Goodwin, & Whittington, 1998), 2) influence
ratings of observed leadership behaviors (Eden & Leviatan, 1975), and 3) determine the
quality of the leader-follower relationship (Engle & Lord, 1997).
One‘s implicit leadership theories are indicative of the amount of leadership
experience accumulated by a person. As one becomes more experienced as a leader, one
should have more elaborate theories about how leaders ought to behave and how
leadership ought to function. Hence, when asked to articulate their implicit leadership
theories, the output from more experienced leaders should be more elaborated and
explicit.
35
H1d: The quality of implicit leadership theories is positively correlated with prior
leadership experience.
Role of self-reflection in transforming leadership experience
This section of the dissertation discusses the main thrust of this research, which
concerns how leadership experience is transformed by reflection. I will introduce the
performance and developmental variables of concern, and build the case for the
mediating role that critical reflection plays between leadership experience and these
outcome variables.
Reflection is useful in terms of its influence with three major aspects of
experience – in grasping experience, in transforming experience, and in resolving
puzzling aspects of new or old experience (sees Figure 2). In the figure, reflection
represents the process by which experience is grasped, transformed and applied to new
puzzles of practice (arrows 1, 2 & 3). At each stage, learning and therefore development
can occur because of reflection. This reflection also occurs after the solution or course of
action has been implemented, because the outcomes of the course of action represent new
experience that needs to be assimilated (arrow 4).
In the terminology of Argyris and Schon (1978), this last stage constitutes double-
looped learning, which has been applied in organizational settings. For example, the US
military recognizes this learning and feedback practice as after-action review (AAR). In
the Israeli military, after-event reviews (AER) are conducted in which actors are required
to provide self-explanations for both successful and failed events that have occurred,
followed by data verification and feedback (Ellis & Davidi, 2005). Although not
referenced directly in the Israeli model, their implementation of organizational learning
36
are generally along the same lines of Dewey‘s grasping and transformation of experience.
In the sample for the study, the AAR process employed as part of the training curriculum
is typically problem-focused, with a strong emphasis placed by the instructors on
problem identification and explanation to trainees on what went wrong.
This dissertation will focus mainly on the transformative action of reflecting on
past experience, though in the following review, the use of reflection in all phases of
experience will be briefly introduced. In addition, for the purpose of the dissertation,
reflection that occurs during action (reflection-in-action) is presumed to occur but will
not be manipulated as part of the study design.
What then is reflection? Reflection is a form of critical thinking. Dewey defined
reflection as a form of thinking that involves ―active, persistent and careful consideration
of any belief or supposed form of knowledge in light of the grounds that support it and
the further consequences to which it leads.‖ (Dewey, 1933, p.9). Such thinking is not
procedural, but holistic, moving beyond mere problem solving to also involving one‘s
intuition, emotion and passion (Greene, 1988). From its early days of conceptualizing
reflection as a better way to encourage learning in educational settings by Dewey, it has
been extended into the domain of professional learning through Schön‘s (1983) work on
the tacit knowledge of practitioners. Schön posited that reflection can occur for an event
that has already occurred (reflection-on-action), for an event that is occurring (reflection-
in-action), or for a future event (reflection-for-action). The focus of this dissertation is on
reflection-on-action, since we are dealing with how past experience gets transformed
through reflection.
37
Typically the events that trigger reflection are ‗puzzles of practice‘ (Munby &
Russell, 1990), which is a difficulty, or something irresolvable that has been encountered.
A problem, as defined by Newell and Simon (1972) has been encountered by one when
―he wants something and does not know immediately what series of actions he can
perform to get it.‖ (p.72). Problems are considered complex when they require ―decision
processes that have not been encountered in quite the same form and for which no
predetermined and explicit set of ordered responses exist‖ (Mintzberg, Raisinghani, &
Theoret, 1976, p.246). However, as will be seen in the review of reflection from a
different discipline later in this dissertation, research in clinical and personality
psychology typically manipulate reflection and ruminative thinking through raising
awareness of self and channeling attention to oneself (e.g. Duval & Wicklund, 1972;
Nolen-Hoeksema, 2000).
Reflection, experience, experiential learning and learning styles
Dewey‘s conceptualization of reflection as a form of thinking useful for learning
had a great influence in shaping later research in learning and experiential learning. For
example, Kolb‘s work on individual differences in learning styles acknowledges its
theoretical debt to Dewey, amongst others such as Piaget and Lewin (D. A. Kolb, 1984).
The core essence of reflection lies in meaning-making. Mezirow, building on Bruner‘s
four modes of meaning-making (Bruner, 1996), claimed that his transformative learning
theory adds a fifth dimension of critical self-awareness of one‘s tacit assumptions and
assessing their relevance for meaning-making (Mezirow, 2000b), of which critical
reflection is an integral component of his theory. This has important implications for
developing leaders. The ability to reflect on experience is evident in leaders who are
38
higher in cognitive development (R. Kegan, 1994; Kuhnert & Lewis, 1987). Such leaders
demonstrate higher levels of self-awareness and critical reflection and openness to their
tacit operating assumptions. From a constructive-developmental framework, Kegan refers
to ‗meaning-forming‘ and ‗reforming our meaning-forming‘ as two critical processes for
one‘s development (Robert Kegan, 2000).
Conceptually, these two processes appear to operate within the same space as
Kolb‘s ‗grasping‘ and ‗transforming‘ of experience (D. A. Kolb, 1984). Kolb approaches
experience and meaning-making from an individual difference perspective. Kolb‘s
experiential learning theory (ELT) proposes two modes of grasping experiences, concrete
experience and abstract conceptualization, and two modes of transforming experience,
namely reflective observation and active experimentation. According to his model,
experiences form the raw inputs for one to reflect on, which again resonates with Schön‘s
concept of reflection-on-action. Insights are then assimilated into abstract concepts and
implications for actions. These outputs can be actively tested and serve to influence the
way subsequent new experiences are processed.
At the risk of oversimplification by way of ignoring the iterative nature of these
processes, one could suggest that experience is first apprehended, or ‗grasped‘ to use
Kolb‘s terminology. Kolb‘s use of the term ‗transforming experience‘ can be elaborated
further by Kegan‘s use of the terms ‗meaning-making‘ and ‗reforming meaning-making.‘
Both sets of terminology share similarities with regards to identifying, dissecting and
evaluating tacit assumptions regarding one‘s experience, which are core tenets of
Mezirow‘s transformative learning theory.
39
From an organizational perspective, the notion of meaning grasping can also be
seen in the work of Argyris and Schön (1978). Single-looped learning can be construed
as a form of meaning-making by which consequences of actions are evaluated against the
initial course of action that produced the consequences. Conversely, double-looped
learning goes back to an evaluation of the initial governing variables that led to the
course of action – a process which appears then to be a form of reformulation of the
meaning-making process. Hence, the previous discussion of multiple levels of
conceptualizing experience applies here, since experience is not just an individual level
construct as initially assumed by Dewey. Rather, as discussed previously, experience in
this instance takes the form of shared discourse, feedback, collective exploration and
meaning-making. The collective nature of experience in this regard is the crux of
organization learning that is discussed by Argyris and Schön (1978).
If one was to adopt a more social-cognitive perspective, reflection and reflexive
action can also be understood when contrasted against routine action (Dewey, 1933).
Routine action occurs when one feels that there is a high degree of certainty and stability
in one‘s life, a term which Walter Mischel describes as a ―strong psychological situation‖
(Mischel, 1973). Under such situations, the cues guiding prescribed behaviors are strong,
for example, when there is strong tradition, authority, or impulse. When problems are
categorized as familiar enough, or when the psychological situation is strong enough, the
principle of least effort (Allport, 1954) or some other satisfysing heuristic (Tversky &
Kahneman, 1974) comes into play and problem solving in such instances typically adhere
to a template or script in some degree. Hence, in order to initiate reflection on past
events, one must provide an impetus for reflection to occur – some form of cues need to
40
be present that causes the recall to deviate from an automated process, to one in which it
gets reconstructed into a different perspective based on these cues.
A related point regarding reflection concerns the boundary conditions for
reflective action, particularly reflection-in-action. Reflection is not useful when there are
good enough pre-existing scripts for actions. This is evidenced by definitions of
reflection reviewed earlier, in which a necessary pre-condition is the presence of a
‗puzzle of practice‘ (Munby & Russell, 1990) or a professional dilemma (Schön, 1983).
From the perspective of self-regulation theories (Duval & Wicklund, 1972; Martin &
Tesser, 1996), reflection is a form of self-directed thought which is initiated when there is
a perceived discrepancy between one‘s current state versus a desired state. When actions
are automated, or discrepancies are not perceived, then reflection is not the default mode
of thought.
This has important implications for encouraging leaders to reflect more on their
leadership experiences. Without sufficient reflection, routine action and non-reflective
responses to perceived routine situations can lead to an over-reliance on pre-defined
scripts. When this occurs, cognitive miserliness is said to have set in (Fiske & Taylor,
1984), bringing along with it biases and errors in judgment that may have serious
leadership performance implications (Moskowitz, 2005). In order for reflection to be
initiated, one must first experience a ‗puzzles of practice‘ (Munby & Russell, 1990), or a
difficulty or something not immediately irresolvable, or perceive a discrepancy between
current versus desired states. These mental states allow one to feel removed from the
constraints of a strong psychological situation. Only when this occurs is one prompted to
step back to analyze their experience. This stepping back for analysis applies to
41
experiences that are currently occurring (reflection-in-action), as well as to experiences
that have already occurred (reflection-on-action) (Schön, 1983), and is critical for
learning (Ellis & Davidi, 2005).
Reflection as a trait and as a learning activity
Thus far, the discussion of the role of reflection in making experience meaningful
has not touched on typologies of reflection uncovered by later research. Two typologies
will be discussed briefly.
Firstly, unlike the current treatment of reflection as a learning process or an
activity, reflective capacity can also be construed as a personality trait. For example,
some scholars regard self-reflectiveness together with internal state awareness to
comprise private self-consciousness (e.g. Cramer, 2000; Creed & Funder, 1998; Nystedt
& Ljungbert, 2002). Others make conceptual distinctions and identify different
psychological and behavioral outcomes between different types of reflection, including
rumination and experiential mindfulness, amongst others (Nolen-Hoeksema, 2000;
Teasdale, 1999; Trapnell & Campbell, 1999). Some of these individual differences can
eventually influence the types of preferences one has with regards to how experience is
grasped, for example as measured by the Learning Style Inventory (LSI) (A. Y. Kolb &
Kolb, 2005). Given that this dissertation adopts a developmental perspective, the activity
of reflection is regarded here as being of greater significance than its trait component,
although the reflective behaviors, regardless of trait or situational causes, share a similar
distinction in terms of adaptive and maladaptive outcomes, as will be reviewed next.
42
Adaptive versus maladaptive forms of reflection
The traits perspective offers a second typology. This pertains to adaptive versus
maladaptive forms of reflection, and is relevant to our treatment of reflection as a
developmental activity. For example, Trapnell and Campbell (1999) distinguished
between rumination versus self-reflection as different forms of self-attentiveness brought
about by different causes. In the former, it is brought about by ―perceived threats, losses,
or injustice to self‖ whereas in the latter it is brought about by ―curiosity or epistemic
interest in the self‖ (Trapnell & Campbell, 1999, pg 297). In depression research,
rumination is defined as ―behavior and thoughts that focus one‘s attention on one‘s
depressive symptoms and on the implications of these symptoms.‖ (Nolen-Hoeksema,
1991, p. 569). Rumination is a maladaptive form of reflection, because it is not only
moderately associated with depression but with anxiety as well (Mor & Winquist, 2002).
Rumination is also maladaptive because of a perseveration component in which
cognitive resources are taken up with persistent dwelling on aspects of a problem, its
symptoms and its causes with no intention towards problem-solving (Nolen-Hoeksema,
1991, 2000). Additionally, rumination is maladaptive when it directs cognitive resources
that would have otherwise been useful for problem-solving into a form of abstraction of
the problem (Teasdale, 1999; Teasdale et al., 2000). This abstraction takes the form of an
extrapolation of the problem in improbable or illogical ways. This is the ―Why me?‖
component of maladaptive reflection. Hence maladaptive reflection is brought about by a
perceived threat in the environment, and manifests itself as a persistence to dwell on a
problem without any effort to engage in problem-solving, or as a tendency to draw from
the problem illogical or improbable implications that further implicates the self.
43
In Trapnell and Campbell‘s (1999) distinction, self-reflection is healthy when it
arises out of intrinsic curiosity and non-judgmental desire to learn about oneself. Watkins
(2004) showed that the non-judgmental mindfulness or openness towards emotional
aspects of a negative experience can facilitate emotional processing and recovery from
that bad experience. Self-regulation theories suggest that non-judgmental openness to
aspects of the self can lead to greater self-awareness and self-knowledge and more
effective self-regulation (C. Carver, S. & Scheier, 1982; C. S. Carver & Scheier, 1990).
When the same mode of thinking is applied to aspects of a problem encountered, these
theories also suggest that problem-solving is facilitated, because they are useful during
the fact-gathering phase prior to problem-solving (Martin & Tesser, 1996; Teasdale,
1999).
One may ask exactly how reflective thought is adaptive. Some research suggests
there may exist different modes of processing which may be (mal)adaptive in different
ways. For example, Teasdale (1999) proposed three mutually exclusive modes of
processing that possess distinctive functional properties. At the sensory level, ‗mindless
emoting‘ focuses purely on the sensory information, is completely absorbed in the
emotional experience. Since this mode contains no conscious reflection at all, it is
irrelevant to the current discussion here and will not be elaborated on further.
The next mode of thinking is a conceptual-evaluative processing of discrepancy
information, focusing on evaluations of discrepancies between current and desired
outcomes (called the comparative-evaluative mode). Note that this form of thinking
easily becomes maladaptive when perseveration sets in, because it then becomes similar
to ruminative thinking. However, some degree of conceptual-evaluative processing is a
44
necessary component of problem-solving, which is a characteristic of adaptive reflective
thinking.
The utility of these modes of thinking also depends in part on the type of
problem-solving being undertaken. Depression research suggest that this mode of
thinking appears to hinder concurrent problem-solving (Watkins & Baracaia, 2002), is
related to brooding and to the onset of bouts of depression in the future (Treynor,
Gonzalez, & Nolen-Hoeksema, 2003).
One possible explanation for its interference with concurrent problem-solving
could be due to the expenditure of conscious effort required to maintain this mode of
thought. This takes away from limited cognitive resources available for the problem
solving task. Also, as shown by the review above, the excessive focus on discrepancies
could be related to future depression. If this is true, then it appears that this mode of
thinking may be maladaptive when used for concurrent problem-solving, but may have
its maladaptive properties mitigated when used in processing of problems that do not
conflict with similar usage of cognitive resources, e.g. such as in evaluations of
performance of past problems that have already occurred.
The third form of thinking in reflection is experiential-awareness, characterized
by non-evaluative, intuitive, in-the-moment awareness of the experience at hand. This
mode of thinking shares many similarities with mindfulness (K. W. Brown & Ryan,
2003), of which the most significant commonality is that both emphasize a non-
judgmental openness to the current experience. This mode of thinking appears best suited
for concurrent problem-solving tasks (Watkins & Baracaia, 2002). One explanation for
this could be due to the greater openness to factual information that may be of subsequent
45
use in problem-solving. A recent experiment concerning the utility of this mode of
thinking on subsequent social-problem solving appears to provide indirect support for
this explanation (Watkins & Moulds, submitted).
Hence, adaptive reflection is motivated by intrinsic curiosity, characterized by an
openness and exploration of the current experience, of staying in the moment, so that the
aim of problem-solving may be successfully attained.
Ways of invoking reflection
The traits perspective of reflection describes a behavioral tendency for people to
engage in reflective thought. However, reflective thinking has been shown to be
amenable to artificial inducement, and is commonly manipulated in psychology research
(e.g. Nolen-Hoeksema, 1991; Watkins, 2004). At the individual level, journaling is one
form of reflection that originated from its historical roots in the educational praxis and
has made its way to leadership development practices (Bolton, 1999; London, 2002).
Above the individual level, reflection is also postulated to occur in social interactions,
such as in dialogues and peer feedback (Malinen, 2000). In one example, Cameron,
Hayes and Wren (2000) utilized community-level reflection to evaluate a nursing and a
physio/occupational programs in a medical setting. The role of dialogue in reflection is
critical – factors that inhibits or promotes communication, such as the communication
medium, in turn impact the quality of communal reflection that occurs (Clyde & Klobas,
2000; Russell & Cohen, 1997).
From a constructive-developmental perspective, leaders who consciously engage
in reflection can be expected to become better leaders (Kuhnert & Lewis, 1987). Such
leaders are able to articulate a higher level of appreciation for the intricacies of
46
leadership. Leaders who actively engage in reflective inquiry are more liable to
demonstrate what Quinn (2005) terms as the ‗fundamental state of leadership.‘ According
to Quinn, it is taxing to remain indefinitely in that state. Yet he believes that leaders can
be assisted to enter that state, simply by asking the following four reflective questions:
1) Am I results centered? (Am I willing to leave my comfort zone to make things
happen?)
2) Am I internally directed? (Am I behaving according to my values rather than
bending to social or political pressures?)
3) Am I other focused? (Am I putting the collective good above my own needs?)
4) Am I externally open? (Am I receptive to outside stimuli that may signal the need
for change?).
From the previous discussion on adaptive reflection, it is clear that the above list of
reflective questions suggested by Quinn is adaptive in nature and helpful for
development, because it contains elements of problem-focused solving and avoids the
ruminative and problem-dwelling components of maladaptive reflection.
Given the above discussion so far, this dissertation advances the following general
hypotheses:
H2a: The relationship between leadership efficacy and leadership experience is
moderated by type of reflection triggers.
H2b: The relationship between leadership self-awareness and leadership
experience is moderated by type of reflection triggers.
H2c: The relationship between quality of implicit leadership theories and
leadership experience is moderated by type of reflection triggers.
47
Experience, reflection and leadership efficacy
Leadership experience is stored in one‘s memory. More specifically, leadership
experience is a form of episodic memory (Tulving, 1972) in which the self is involved in
some way as the main leadership actor or part of the cast of actors in the leadership
experience. Stored as part of the experience is semantic information concerning oneself
as a leader or as part of the leadership process. Both types of information are part of one‘s
declarative knowledge (Winograd, 1972), which is our statements of beliefs about people
and things that contribute to our understanding of ourselves and how we relate to and
function in the world.
Memory, and more specifically, declarative knowledge, is based on cues for either
retrieval or reconstruction, depending on which view of memory one subscribes to
(Smith, 1998). Information pertaining to the self that gets recalled along with details
about an event can be perception-based (images of oneself) or meaning-based (i.e.
organized around ideas or theories one has about oneself) (Kihlstrom & Klein, 1994).
Either way, one‘s ideas about oneself is intricately linked to one‘s memory for oneself
(Klein, 2001).
Hence, with regard to leadership experience, how these experiences get recalled
or reconstructed (and the self-information that gets recalled along with the experience)
may result in a qualitatively different effect on the self. For example, recalling a
leadership experience in which one‘s semantic information about oneself as a leader is
mostly negative may have adverse impact on one‘s subsequent perceptions of oneself as a
leader. Given the previous discussion of how efficacy beliefs are a form of self-
evaluation, efficacy beliefs can be influenced by these perceptions because they feed into
48
one‘s own evaluations of oneself as a leader. Hence, efficacy beliefs are influenced by the
type of reflection that occurs because reflection influences the type of self-information
that is made salient. This self-information is then utilized during self-evaluation, which
subsequently influences one‘s efficacy beliefs.
Conversely, if the cues for recall are tweaked, such that from the same leadership
experience, only selected bits about oneself as a leader are utilized in the reconstruction
of the leadership episode, then the negative effect of the leadership experience on one‘s
perceptions of oneself as a leader may become attenuated, or less severe. Going a step
further, one can be prompted to reflect positively on the recall of the experience, such
that the beneficial effects get even more accentuated. Hence, reflection can work together
with prompted recall to manipulate the way individuals attend to different aspects of
one‘s leadership experience such that the event gets reinterpreted in adaptive or
maladaptive ways. Furthermore, one would expect that when recall and reflection are
manipulated in the same direction (i.e. recall and reflect on positive aspects of leadership
experience), its impact would be greatest.
In this dissertation, due to constraints on the current design, this mode of inquiry
will be pursued in which type of recall is always in the same direction of type of
reflection being introduced. In addition, the type of leadership experiences under
investigation will also be controlled for. These will be common military training
experiences. Such training experiences will include being assigned leadership
responsibilities and appointments in military exercises. High leadership experience would
refer to higher amount of exposure to these common military training activities
49
undergone. Hence, combining the review on adaptive and maladaptive reflection together
with this section‘s review on efficacy, I propose that:
H3a: Recalling and reflecting positively on any personal leadership experience
increases subsequent leadership efficacy.
H3b: Recalling and reflecting negatively on any personal leadership experience
decreases subsequent leadership efficacy.
Reflection and Leadership Self-Awareness
There is a wide range of definitions and operationalizations of the term ―self-
awareness‖, but the common intuitive meaning of the term involves using some concept
of oneself to affect one‘s own behavior (Mulhauser, 1998). Unfortunately, such a broad
definition has led to a variety of operational differences and domain specific
interpretations. For example, a major stream of research in psychology conceptualized
self-awareness as focusing attention on oneself (Duval & Wicklund, 1972; Silvia &
Duval, 2001; Wicklund, 1979). Yet such a conceptualization of self-awareness is vastly
different from the content knowledge-based conceptualization of self-awareness as
‗knowing oneself‘ (Harter, 2002) used by leadership practitioners. Consequently,
different effects of self-awareness have been hypothesized. For example, being more self-
aware in one sense of the word (i.e. knowing oneself better) is hypothesized to lead to
greater perspective-taking, higher ethical conduct, balanced processing, higher levels of
development and trust in followers (Gardner, Avolio, Luthans, May, & Walumbwa,
2005).
On the other hand, being more self-aware as defined by some psychologists (i.e.
focusing attention on oneself) has at times been cast in a negative light, particularly in
50
some clinical research (for a review, see Ingram, 1990). Some clinical psychology
research have demonstrated self-awareness can be associated with an increase of
depressive episodes in depression patients (Pyszczynski & Greenberg, 1987), is
correlated with negative affect and reduced well-being in a meta-analytic review (Fejfar
& Hoyle, 2000), and has been discussed in the same vein with socially irresponsible
behavior (Prentice-Dunn & Rogers, 1989).
In any case, two common elements to the above definitions are that self-
awareness either proceeds after or include some form of self-examination, and that the
resultant outcome is some form of knowledge change, leading to behavioral regulation.
These elements are also processes by which self-reflection is defined. Hence, regardless
of which operational meaning of self-awareness is used, it is clear that self-reflection will
always be associated with heightened self-awareness. Indeed, self-awareness in leaders is
synonymous with personal insight in leaders, with self-reflection and introspection
(Dewey, 1933; Schön, 1983) as operating processes through which leaders gain clarity
and concordance with respect to their core values, identity, emotions, motives and goals
(Avolio, Gardner, Walumbwa, Luthans, & May, 2004; Avolio & Luthans, 2006; Gardner
et al., 2005; Luthans & Avolio, 2003). Shamir and Eilam (2005) posited that leaders who
are true to themselves need to be in touch with their core essence, and be clear about what
defines them as people and as leaders, that is more self-aware.
Given this, I advance that:
H4a: Recalling and reflecting on past experience results in greater self-
awareness.
51
Does self-awareness increase depending on the type of self-examination? Earlier,
a distinction was made between more adaptive versus maladaptive types of self-
examinations. In the former, the self-examination is characterized by focus on problem-
solving, while with the more maladaptive type one is more ruminative, more self-focused
and dwelling more on the symptoms of the problem rather than on solving the problem.
Hence, it is likely that the maladaptive types of self-examinations should result in greater
self-awareness rather than the more adaptive types, given the larger focus on the self.
Given this, I advance:
H4b: Ruminative recall of past experience results in greater self-awareness
compared with adaptive recall.
Reflection and Implicit Theories of Leadership
How does reflection influence the nature of accumulated experiences? Does
reflecting on experience provide incremental improvements in one‘s implicit leadership
theories? Building upon the arguments advanced in the previous section regarding the
role that reflection has on greater elaboration in one‘s life story and leadership
development, one should also expect reflection to similarly have a positive effect on
one‘s implicit leadership theories.
In addition, because reflection serves a function of making explicit the implicit
assumptions, any form of reflection should result in a more elaborate implicit leadership
theory being explicated than no reflection at all. However, it is unclear if adaptive and
maladaptive reflection exerts differential effects on one‘s implicit leadership theories.
The argument for adaptive reflection having a greater positive effect on implicit
leadership theories hinge on the problem-solving nature of adaptive reflection, which
52
should hypothetically lead to greater clarity of cause-effect linkages in one‘s implicit
leadership theories. On the other hand, ruminative reflection, being by definition focused
on one‘s self, may also trigger greater clarity about one‘s leadership identity, hence
resulting in one‘s implicit leadership theories being impacted.
Evidence from clinical research in depression suggests that maladaptive reflection
is more strongly associated with moods and affect, while any form of rumination results
in greater elaboration of one‘s internal state awareness (Nolen-Hoeksema, 2000; Silvia,
2002). Hence, the type of explication of one‘s implicit leadership theory may well depend
on the nature of reflection that occurs.
Adaptive reflection may result in an explication of implicit leadership theories
that are more positive and problem-solving focused, while maladaptive reflection may
lead to explications of implicit leadership theories that are more negative and problem-
aversive. In other words, the type of reflection primes or sensitizes related aspects of
one‘s implicit leadership theories, i.e. through spreading activation along an associative
network (Cantor & Mischel, 1977; Neely, 1977). When this occurs, these aspects of one‘s
implicit leadership theories become more responsive to environmental triggers, such that
they attain heightened accessibility (Meyer & Schvandeveldt, 1971). Hence, following
reflection, one should expect that when a leader articulates his/her implicit leadership
theory, the output should be dependent on the type of reflection that has occurred.
The priming effect of reflection on heightening accessibility of various aspects of
one‘s implicit leadership theory should also have an impact on the latency response to
cue words. This relationship between latency response and priming is even more
pronounced when the prime is affect-laden, or highly emotive (Niedenthal, Halberstadt,
53
& Setterlund, 1997). Likewise, one should expect that when leaders are tasked to make a
timed judgment on whether cue words describing leadership are positive or negative in
nature, latency response times for cue words that are positive should be shorter than
negative words following adaptive reflection. The reason for this is because the type of
reflection that has occurred before the task serves as a prime for this lexical decision task,
thereby inducing heightened accessibility in different aspects of one‘s implicit leadership
theory.
Regardless of the valence, both types of reflection should lead to more complex
and explicit theories of leadership, though it is unclear between the two types of
reflection which would have the greater effect. Hence, the following hypothesis posits a
more global effect of reflection on elaboration of one‘s implicit leadership theories, with
no further predictions advanced on potential differential effects of different types of
reflection:
H5a: Adaptive reflection results in positive aspects of one’s implicit leadership
theories becoming more easily explicated.
H5b: Maladaptive reflection results in negative aspects of one’s implicit
leadership theories becoming more easily explicated.
Summary and integration of literature
This dissertation proposes that different forms of reflection on one‘s leadership
experience can have differential effects on one‘s leadership efficacy, implicit leadership
theories and subsequent leadership performance. Reflecting on a leadership episode
occurs at the point of recall of the episode. In this dissertation, both recall and reflection
are paired for adaptive or maladaptive purposes. Maladaptive recall and reflection
54
focuses on the nature of problems encountered, amplifies the implications of problems,
and generally causes one to dwell on problems with no intention of moving towards
problem-solving. Adaptive recall and reflection, on the other hand, is characterized by
experiential mindfulness, which involves a state of suspended judgment while remaining
open to allow salient aspects of a problem to emerge. It is focused on problem solving
rather than merely ruminating on a problem.
This dissertation postulates that when reflection is adaptive, one‘s leadership
efficacy is raised, one‘s implicit leadership theories are better explicated and more
developed, with positive aspects of one‘s implicit leadership theories more accessible.
When reflection is maladaptive, the processing that occurred may create negative
emotional reactions and self-attributions that lower leadership efficacy. When one
engages in maladaptive reflection, one‘s implicit leadership theories are also developed,
but the negatively valenced aspects of one‘s implicit leadership theories are also made
more accessible.
55
CHAPTER 3: STUDY DESIGN AND METHODOLOGY
56
Chapter 3: Study Design and Methodology
This chapter provides an overview of the study methodology that was
implemented, including information on the participants, study design, procedures and
manipulations, and measures.
Research Design
A three-condition between subjects design was used to evaluate the effect of
invoking various forms of self-reflection on one‘s leadership efficacy, Psycap and
implicit theories of leadership.
Sampling Plan and Power Analysis.
A priori power analysis conducted to estimate the sample size needed to obtain
the desired effects revealed that a minimum of 207 participants were needed. Power
analysis examines the relationship between four variables: the size of the effect (r), the
sample size (N), the type I error rate (α), and the type II error rate (β) (Freedman, 1982).
First, given the relationships seen in the past, a very conservative estimate of the effect
size of about .25 was anticipated for each condition. Next, the type I error rate (α) of .05
was utilized for the calculation. Last, the type II error rate (β) is set at .10 so 1-β= .90.
Given these figures, the freeware GPOWER software program (Erdfelder, Faul, &
Buchner, 1996) was used to derive that the sample size required from this power analysis
was 69 participants per group. Given that this is a 1x3 study design with 3 cells, the
minimum participant required is 207.
57
Participants and Training Context
321 conscript junior commander trainees organized into 8 training platoons in 2
companies from a junior military leadership school were tracked over an 8-week
leadership course. Participants were 18-year old Asian males identified to undergo
leadership training as part of their conscription requirement. As part of the leadership
potential shortlist process, these trainees have a minimum of junior college/polytechnic
education (the US equivalent of senior high school) and demonstrated good leadership
potential and leadership performance.
These soldiers have previously undergone 9 weeks of military socialization and
basic military training. During the 9 weeks, these soldiers are introduced to basic infantry
field craft, fitness, regimentation, weapon handling and shooting. Following the 9-week
training, trainees with identified leadership potential embark on an 8-week leadership
course which comprise experiential training in military domains such as further weapons
training, tactics and signals as well as introduction to leadership. This training prepares
them for operational roles leading platoons of men. As part of their development, trainees
are introduced to experiential learning techniques such as journaling, conducting and
participating in action reviews (AARs). Implicit to these techniques is the expectation
that trainees will exercise reflection throughout the training course, although how
reflection is actually carried out by the trainees is not monitored by the training school.
Pilot Studies
Two pilot studies were conducted. The first pilot was a qualitative study utilizing
randomly selected trainees from a more senior cohort of the same leadership training
58
course. The second pilot was a quantitative study to establish the reliability of all
measures used.
Pilot Study 1. This qualitative study had two aims. Firstly, it aimed to establish
the type and nature of salient leadership events in the formal training curriculum that
generated leadership developmental triggers for trainees. Secondly, from the reflections
around these developmental triggers by the trainees, the study aimed to establish the
leadership-related content that the reflections revolved around so as to serve as a bottom-
up approach towards creating a scale to measure the quality of reflection that occurred
during the leadership training course.
A total of 10 trainees were randomly selected from a graduating cohort of the 8-
week leadership training course. Each trainee was interviewed by a research intern naïve
to the study objective in a one-to-one setting. The research intern had an alternate
research objective to understand and collect stories on military socialization that occurred
in the training course.
During the interview, each trainee was asked to record his recall of salient events
that impacted his leadership development on an event history calendar chart. The
interviewer then probed the trainee regarding each event on the calendar for reasons why
the event was viewed as impactful. Where the trainee felt that he had learnt something
about leadership or about himself as a leader, the researcher registered a trigger as having
occurred. Each event may generate more than one developmental trigger. Table 1
tabulates the key events and the percentage of developmental triggers reported by the 10
trainees. From this study, the two events in the training course that generated the highest
59
proportion of developmental triggers were used as the focal developmental activities in
subsequent studies for reflection.
------------------------------------------------------------
Insert Table 1 Here
------------------------------------------------------------
Interestingly, for all 10 trainees, all the salient events that were identified as
developmentally impactful were tough, somewhat unpleasant events. Examples of such
themes included the hardship they suffered or the harsh treatment inflicted on them by
their peers or superiors during the field camp. Other examples involved being pushed /
pressured to continue beyond one‘s limit of endurance in a route march exercise,
experiencing team conflict and role uncertainties during navigation exercise, being given
poor feedback in AARs, or even seeing undesirable leadership behaviors in peers and
instructors that trainees resolved not to role-model after. Significantly, there were no
positive triggers (e.g. instructor or peer exhibiting exemplary leadership) identified by the
trainees. Upon consulting with the commander of the training school, he concurred that
training was extremely onerous for the trainees.
As the overall tone of the training environment was quite negative, there were
initial concerns that any form of cognitive processing about the events may actually do
more harm than good. The fear was that the negative events may result in trainees tending
to ruminate rather than reflect positively. In discussing with the trainees from the pilot
study, all of them unanimously agreed though that the leadership course was overall
beneficial for them. As such, in spite of concerns that reflecting on the events from this
60
leadership course may trigger negative emotional reactance, I decided to proceed with the
study.
As mentioned, a second aim of the study was to establish the nature of the
leadership-related content of the reflections that occurred as result of the developmental
triggers. These triggers were derived from a thematic analysis of the reflections told by
the trainees about their experiences. These reflections were either volunteered by the
trainees, or were prompted for by the interviewer as the trainees were relating their
stories and experiences. To guide the interviewer, a set of suggested prompts were
provided to the interviewer. This is found in Appendix C.
All in all, nearly 5 hours of interview material were recorded by the interviewer.
Of these less than 20 min of the recording were usable leadership-related reflections – the
remainder of the recordings made were narrations of stories, pauses, clarifications and
off-tangent topics. Bearing in mind that the interviewer was naïve to the intent of the
study the interviewer was asked what impressions she had about the trainees‘
socialization experiences. Her main impressions revolved around how trainees became
more aware of themselves, and how trainees became better over time. From this, I
derived that self-awareness and self-regulation were two themes that may be pertinent to
the reflections by the trainees. In addition, the intern also identified that the most salient
characteristics of the leadership-related content from which the trainees either became
more self-aware or more self-regulated had to do with the trainees‘ values, strengths or
weaknesses.
From discussion with the interviewer, I derived a coding frame for evaluating the
20 min of recorded reflections obtained from this pilot study. The percentage of reflection
61
statements captured by this coding frame from all the reflection available is about 63%
(See Table 2 below). This coding scheme also contributed to the creation of a set of items
to measure the quality of reflection outcomes that will be used in the actual study.
------------------------------------------------------------
Insert Table 2 Here
------------------------------------------------------------
Scale-creation: Quality of Reflection Outcome Scale
The qualitative study helped identified the nature and focus of reflections that
were feasible from the leadership training course. Specifically, it identified that about two
thirds of the reflections can be classified as reflections that pertain to being more aware
or more self-regulating of one‘s values, purpose, leader self-image, operating paradigms
and assumptions. Since these reflections were triggered by salient events available in the
leadership training course, it meant that one can expect trainees to, at some point during
the course, be able to provide some level of reflection around 1) his values as a leader, 2)
his image of himself as a leader, and 3) his leadership operating paradigms and
assumptions. In addition, these reflections could also demonstrate greater self-awareness
or greater self-regulation around these three domains.
Given this information, five items per construct were created as an initial bucket
pool for a total of 30 items. These 30-items were then subjected to a bucket drill exercise.
2 assistants naïve to the study were separately assigned to sort the 30-items into the 3 by
2 coding frame. Of these, 19 items were correctly assigned to the coding frame by both
assistants. The rest of the items were mis-assigned by either one of the assistants. None of
the items were completely mis-assigned by both assistants. The 11 mis-assigned items
62
were then re-evaluated by the assistants together. They achieved agreement on 2 of them.
These 2 items were then added back into the pool for a final item count of 21. The next
step was to conduct an exploratory factor analysis (EFA) to better understand how the
items will perform empirically.
Pilot Study 2. The second pilot study involved an EFA for the newly created
scale and confirmatory factor analyses (CFA) to examine the reliabilities of the rest of the
measures used. This quantitative study utilized 497 trainees from a basic military training
school. These trainees are emerging adults, all males aged 18 who are compulsory
conscripted enlistees. Their racial composition is consistent with national averages, which
are 70% Chinese, 20% Malays and the remainder from other racial groups. This
composition is also consistent with the actual study sample to be used later. At time of
survey administration, these trainees are approximately midway through their 9-week
basic military training and socialization program. Top performing trainees from this
program are selected for the leadership training course in the actual study. However, the
actual study utilized leadership trainees from 2 cohorts later, thereby avoiding the
possibility of trainees having prior exposure to the measures used.
Missing data
The survey was conducted via paper-pencil. As such, it was inevitable that despite
on-site checks, there would still be missing entries during survey administration, thereby
resulting in missing data. On the whole, the amount of missing data was less than 5% of
the entire sample. Of these, 3 participants‘ had entries that were completely missing. This
missing data was interpreted as those participants had no interest in completing the
survey. As such, the missing data generated by those participants fell into the category of
63
systematic missing data termed as missing not at random (NMAR) (Rubin, 1976). As this
represented a systematic loss of data, I elected to delete their entries entirely from the
database. With the rest of the missing data, there were no discernible patterns to help
interpret the reasons for missing data in the dataset. As such, the rest of the data was
treated as missing at random (MAR). This missing data was then left to be addressed
using maximum likelihood procedures available in the MPLUS software (Muthen &
Muhen, 1998). To prevent missing data due to incomplete entries, online surveys were
subsequently used with logical checks built in that did not allow for missing entries.
The issue of outliers and normality of the pilot sample was also examined. Details
of how skewness and kurtosis as well as how outliers were dealt with can be found in the
next chapter. These issues had to be addressed in the pilot sample so that the EFA for the
new measure and CFAs on the rest of the measures could be properly executed.
Exploratory Factor Analyses. An EFA, like confirmatory factor analysis
(CFA), is a method to achieve data reduction parsimoniously. Both are multivariate
statistical methods for data reduction for determining the number and nature of common
latent factors that will parsimoniously explain the patterns of observed correlations in
measured variables (Farbrigar, Wegener, MacCallum, & Strahan, 1999). However, EFA
is used for generating theory whereas CFA is used for testing theory. The outcome of an
EFA is agnostic to a priori hypotheses that a researcher has in mind. Rather, it is the sole
function of the ―mechanics and mathematics of the method.‖ (Kieffer, 1999, pg 77).
CFA, on the other hand, cannot proceed without direct intervention from the researcher
hypothesizing the underlying structure of the data at hand. As such, when a new scale is
64
developed, EFA followed by CFA is an essential workflow towards establishing the
validity of the scale.
There are several key decisions involved in an EFA. Firstly, the researcher has to
determine the number of latent factors that best represent the underlying structure of the
data. This is an important determination as over or under-specifying the number of latent
factors can result in the introduction of error in significance testing, factor loading
reproduction and interpretation (Velicer, Eaton, & Fava, 2000).
Scree plot tests (Catell, 1966), setting the minimum eigen (or Kaiser) values
above 1, also known as the K1 rule (Kaiser, 1960), parallel analysis or PA (Horn, 1965)
and the use of latent model fitting techniques with maximum likelihood estimation such
as those implemented in structural equations modeling software like MPLUS (Muthen &
Muhen, 1998) are some ways to assist the researcher in making this determination.
Each method has its weaknesses. For example, the scree plot test is useful and
visually intuitive when the latent factors are unambiguous and strong. In instances with
multiple weak latent factors, the determination of the cut-off based on the plot becomes
increasingly subjective. On the other hand, the K1 rule, though easy to implement,
particularly with the use of statistics software packages, has been consistently shown to
inaccurately over factor (see Glorfeld, 1995 for overview) to such an extent that some
have recommended for its use to be discontinued (e.g. Zwick & Velicer, 1986).
PA involves the use of multiple random generations of datasets with the same
dimensions (sample size and number of measured variables) as the original dataset to
extract (via principal axis or principal components analysis) and order the eigenvalues, so
that the mean and 95th
percentile of all eigenvalues created can be determined. Several
65
reviews such as Glorfeld (1995) and Zwick and Velicer (1986) have found PA to be more
accurate than the use of the scree plot test or the K1 rule.
In this study, a combination of methods was used including scree plots and the
use of maximum likelihood estimation via the MPLUS software to determine the factor
structure.
------------------------------------------------------------
Insert Table 3 Here
------------------------------------------------------------
Table 3 displays the number of latent factors and the extent to which the data
covariance matrix is replicated. From the table, at every level of k, where k is the number
of latent factors, the chi-sq is significant. This means that a k+1 latent factor solution is
feasible. At the 5-factor solution, the chi-sq is still significant. However, no convergence
was possible for a 6-factor solution, which means that 5 latent factors is the highest
number possible with the dataset.
Having established the number of latent factors possible, the next step was to
demonstrate that a 5-factor solution significantly explains the data better than a k-1
solution, i.e. a 4-factor solution. This is done by conducting a chi-sq difference test. From
the table, the chi-sq difference test was highly significant, which meant that a 5-factor
solution offered significant variance explained over a 4-factor solution. Finally, to
establish whether the 5-factor solution was a good fit, its RMSEA value needed to be
compared. According to (Hu & Bentler, 1999), RMSEA values below .06 indicate
satisfactory model fit. In this instance, the 5-factor solution had an acceptable RMSEA
66
value of .044, which indicated that it adequately explained the data generated in the pilot
study.
The final step in the EFA was to explore the item loadings to each latent factor to
determine if the pattern of loadings are aligned with what was intended when the items
were created.
------------------------------------------------------------
Insert Table 4 Here
------------------------------------------------------------
Table 4 shows that most of the items cluster together as expected. The only ill-
performing items were Item 6 which failed to load sufficiently on latent factor 4, or on
any latent factor using the norm of .3 as a cut-off, as well as Item 7 which did not load on
its intended latent factor 5 but loaded on latent factor 2 instead. In addition, Items 10 and
16 also demonstrated tendencies to cross-load on one other latent factor other over and
above its expected latent factor. All in all, the EFA demonstrated that the 21-item scale
appeared to perform sufficiently well as expected such that a CFA using another sample
is justifiable. Hence, I retained the items to test on a more diverse sample.
Confirmatory Factory Analyses. As with subsequent CFAs described
throughout other sections of this dissertation, a combinatory rule composed of three fit
indices was used to determine the goodness of fit of the CFA model. Firstly, the absolute
fit index called the standardized root mean square residual (SRMR) was computed.
SRMR values range from zero to one, with zero indicating perfect global fit of the model.
Secondly, the comparative fit index (CFI) was computed. CFI compares the target model
with a baseline fully identified model. When the target model is as fully identified as the
67
baseline model, the standardized fit is perfect (i.e. attains a value of 1). The aim of the
comparison is to establish parsimony of the target model while maintaining an acceptable
level of fit, which is recommended to be no less than 0.95. Third, the root mean square
error of approximation (RMSEA) was computed by assessing the degree of population
misfit per degrees of freedom, with values closer to zero reflecting better fit and values
around 0.05 representing ‗close fit.‘ Therefore, using a combinatory rule for maximum
likelihood, a Standardized Root Mean Square Residual (SRMR) less than or equal to .08
and either a CFI greater than or equal to .95 or a RMSEA less than or equal to .06
indicates good fit.
Additionally, where required, the CFA models may be tweaked to improve the fit.
This is an iterative process involving one or more of the following steps. To begin, the
zero-order correlation of individual items to the latent factor was examined. Items that
had correlations lower than .60 were evaluated further on two counts, namely – would it
be theoretically justifiable to remove the item from the proposed scale and would
removing the item from the scale greatly improve the overall fit of the scale?
Of the remaining items, should further tweaking be required, the modification
indices provided by the MPLUS output were used. Modification indices, or Lagrange‘s
modifiers, provide an indicator of expected change to the overall model fit should the
recommended loadings (e.g. cross-loading to another latent factor or being allowed to
correlate with other indicators) be freed. In determining whether such cross-loadings
should be allowed, first theoretical fit was used as the primary criteria followed by the
utility at the time, i.e. whether allowing for the cross-loading would result in a better fit
for the sample. For example, in instances where the modification indices recommend
68
that items on the same subscale be allowed to cross-load on each other, cross-loadings
were allowed to occur because it is theoretically justifiable. However, even then, the steps
of lifting the restrictions were conducted one at a time.
A more considered judgment was required when items from different subscales or
even different latent factors were recommended to correlate with each other. These
misspecifications that pertain to the latent factors are reflected in poor SRMR fit indices
(Hu and Bentler, 1999). In such instances, one first checks to see if a similar poor fit was
also apparent across different samples. Here, the modification indices were used based on
the criteria that large indices with small EPCs are indicative of relationships that are
particularly susceptible to sample fluctuations.
Next, the actual wording of the items was closely examined to determine if the
items could have been misinterpreted. In instances where there were sufficient items, the
item was deleted from future analyses to avoid introducing unexplained variances into the
models. In cases where there were insufficient items, an item was retained based on
considering whether it was theoretically justifiable to allow for the cross-loading to
occur.
Measures
New Measure: Quality of Reflection Outcomes. A 21-item survey newly
created as described in the previous section was used here to measure the extent of leader
self-awareness and self-regulation around one‘s purpose and values, one‘s leader self-
portrait and one‘s leadership operating paradigms and assumptions. A higher order latent
factor reflecting the quality of the reflections around these 1st order factors were
69
postulated. In a smaller pilot study, a CFA of the full 21-item scale did not achieve an
acceptable fit. Upon refinement of the items according to the procedure described in the
previous section, the reduced 15-item scale achieved an acceptable fit (SRMR = 0.04, CFI
= 0.96, RMSEA = 0.04). This factor structure was again replicated with acceptable fit in
the actual study sample at Time 1 (SRMR = 0.04, CFI = 0.97, RMSEA = 0.03) and at
Time 2 (SRMR = 0.04, CFI = 0.95, RMSEA = 0.05). The reliabilities for this revised
scale were Cronbach‘s alpha (α) = .849 (for pilot), α=.801 (for T1) and α=.846 (for T2).
The final set of items for this scale, as well as for all other measures used in the
dissertation, can be found in Appendix B.
Generalized Leadership Efficacy. This is a 22-item experimental scale
currently under development by Hannah and colleagues (Hannah, Avolio, Chan, &
Mhatre, 2005). The scale was designed to measure a leader‘s appropriation of his or her
role and environment (agency), and the self-schematic efficacy beliefs (confidence) in his
or her perceived leadership capabilities to organize the positive psychological
capabilities, motivation, means and courses of action required to attain effective,
sustainable performance across their leadership domain.
A CFA conducted on the original 22-item scale did not converge on the
hypothesized structure using the pilot sample. To achieve convergence, 5 items that did
not load highly on any of the latent factors were removed. In addition, some items within
the means and action-transformational subscale were allowed to co-vary. This resulted in
convergence and adequate fit of (SRMR = 0.044, CFI = 0.941, RMSEA = 0.075). Its
reliability was α=.944. The CFI could have been improved marginally by allowing the
subscales of action-means and action-transformational to co-vary. In practical terms,
70
allowing them to collapse into a single latent factor could have been justified given that
that these two latent factors had a very high zero-order correlation of .949. However, I
elected to allow these 2 subscales to remain distinct and retested the CFA on the actual
study sample before deciding.
The results of the CFAs on the actual study sample demonstrated similar adequate
fit at T1 (SRMR = 0.042, CFI = 0.944, RMSEA = 0.064) and good fit at T2 (SRMR =
0.035, CFI = 0.960, RMSEA = 0.059). The reliability of the scale remained high at
α=.908 for T1 and α=.926 for T2. On the basis of these results, the reduced 17-item scale
was retained.
Psychological Capital (Psycap). In the pilot study, CFAs conducted on Psycap
were initially not found to fit well. 6 of the items which fit well were particularly
sensitive to sample fluctuations (i.e. big modification indices with small EPCs) and were
removed from the scale and re-analyzed on the pilot sample with acceptable fit (SRMR =
0.038, CFI = .955, RMSEA = 0.053). In addition, the newly reduced scale was also tested
on the actual study and found to be adequate across Time 1 (SRMR = 0.054, CFI = .948,
RMSEA = 0.049) and at Time 2 (SRMR = 0.051, CFI = 0.963, RMSEA = 0.049).
Overall, the scale reliabilities for the final 18-item set were α=.915 for the pilot sample,
α=.880 at Time 1 and α=.906 at Time 2.
Trait Reflectiveness. The measure used is the Self-reflection and Insight Scale
(Grant, Franklin, & Langford, 2002). This 20-item scale has 12 items measuring self-
reflection ("the inspection and evaluation of one's thoughts, feelings and behavior") of
which half measured one‘s need for self-reflection while the other half measured actual
engagement in self-reflection and 8 items measuring insight ("the clarity of understanding
71
of one's thoughts, feelings, and behavior"; Grant et al., 2002, p. 821). Additionally, 7 of
the items were negatively worded and had to be reverse-coded as instructed by the
authors. It should be noted that one of the items in the scale appeared to be double-
barreled (―I‘m often aware that I‘m having a feeling, but I often don‘t quite know what‖).
A CFA conducted based on the 3-factor model validated by the Grant et al. (2002)
could not converge. Removal of the 7 negatively worded items allowed the model to
converge but the CFA model could not be relied on as the model was non-positive
definite. When technical output was requested from the MPLUS program (via the Tech4
subcommand), two of the latent factors (need for reflection and engaging in reflection)
were discovered to have correlations greater than 1. This indicated that the two latent
factors were so highly correlated that the program could not reliably distinguish between
the two. These two latent factors were combined as it made sense that one with high
need for reflection should also engage in reflection. However, though the CFA ran
successfully with this reduced model, it did not achieve satisfactory fit with the pilot
study sample (SRMR = 0.061, CFI = 0.892, RMSEA = 0.096).
Given the difficulty in trying to replicate the 3-factor model proposed by the
original authors using the current samples, I explored the possibility that participants may
have been responding uniquely to these items in the pilot sample. To do so, an EFA was
conducted with on the pilot sample data.
------------------------------------------------------------
Insert Table 5 Here
------------------------------------------------------------
72
As can be observed from the table, the Insight subscale matched perfectly in the
EFA with the CFA. It was the Need for Reflection and Engagement in Reflection
subscales that were not behaving according to the theoretical structure specified by the
authors. In the EFA, these 2 subscales were observed to cluster into 3 latent factors.
Secondly, there were 3 items (Item 77 ―I frequently examine my feelings‖, Item 82 ―I
have a definite need to understand the way that my mind works‖ and Item 86 ―I often
think about the way I feel about things‖) that were cross-loading across the other 3
unidentified latent factor. Deconstructing these 3 items, I noted that there were themes of
need for reflection (Item 82) and actual engagement in reflection (Item 77) as
hypothesized by the authors. However, there were additional subthemes pertaining to the
content of the reflection (i.e. emotions and feelings, from Item 86) as well as theorizing
about how one‘s mind works (Item 82).
Using the above analysis as an initial frame, connections between items that
clustered together were explored next around the latent factors in an attempt to identify
the nature of these latent factors. It appeared that that F2 captured the essence of both a
need for reflection and the actual engagement in reflection. F2 constituted the need for
reflection and engagement in reflection scales as proposed by the authors were
corroborated by the fact that it contained the largest number of items loading to it.
Looking at the items that loaded highest on F3 (―It is important to me to be able to
understand how my thoughts arise‖) and F4 (―I frequently examine my feelings‖), it
appeared that the F3 had to do with theorizing about how one‘s mind works and F4 had
to do with the reflecting on feelings.
73
From this pilot study, there was partial success in achieving a factor structure for
the trait reflection scale. On the one hand, the insight, need and engagement components
in the scale all appeared as proposed by the authors, but with the need and engagement
components tending to collapse together. In addition, 2 other factors that pertained to the
emotive content of the reflection and the need for theorizing about how one thinks also
emerged.
My main take-aways from this EFA was that the insight component was stable,
while the need for reflection and actual engagement in reflection component should be
merged. Given this partial success, the CFA was repeated with both the actual study
sample at T1 and T2, but choosing this time to merge the need for reflection and actual
engagement in reflection components as one.
Results of the CFAs with these modifications achieved good fit at Time 1 (SRMR
= 0.057, CFI = .960, RMSEA = 0.058) and excellent fit at Time 2 (SRMR = 0.034, CFI =
0.978, RMSEA = 0.046). Given this, subsequent usage of the trait reflection scale at T1
as a covariate utilized the modification of collapsing the Need for reflection with
Engagement in reflection. The Cronbach alphas were α=.842 for the pilot sample, α=.847
for the actual sample at Time 1 and α=.861 at Time 2.
General Self Efficacy. The New General Self-Efficacy Scale (Chen, Gully, &
Eden, 2001) was used to measure general self-efficacy. According to the authors, this 8-
item scale has high content validity, being distinct from self-esteem and uni-dimensional
in capturing self-efficacy. It was also reported by the authors as having high predictive
validity in motivation and work performance and stability across at least two cultural
74
contexts. Its internal reliability were reported by the authors to be at least α = .85 and
above.
In the pilot study, the CFA conducted on generalized self efficacy did not fit well
initially (SRMR = 0.300, CFI = .915, RMSEA = 0.144). Upon removal of 2 items (item
88 and item 94) that were sensitive to sample fluctuations, the reduced scale achieved
acceptable fit (SRMR = 0.024, CFI = .976, RMSEA = 0.093). In addition, the newly
reduced scale was also tested on the actual study sample and its fit was found to be
acceptable as well at Time 1 (SRMR = 0.029, CFI = .965, RMSEA = 0.106) and Time 2
(SRMR = 0.029, CFI = 0.962, RMSEA = 0.126), having satisfied Bollen‘s combinatory
rule of either RMSEA or CFI in combination with SRMR within the respective
acceptable ranges. The scale could have been further tweaked to improve the RMSEA
criteria for model fit. According to Hu and Bentler (1998), the RMSEA is sensitive to
omitted loadings. As such, by allowing for cross-loading between two items (item 95
with item 93 from the pilot sample) the model fit improved to SRMR = 0.018, CFI =
.989, and more crucially, the RMSEA improved to 0.018. However, including these
modifications would have been at the expense of parsimony in the model, which was
deemed unnecessary given the use of the scale as a covariate. The final reliabilities of the
reduced scale were α=.898 for the pilot sample, α=.879 for the actual sample at Time 1
and α=.911 at Time 2.
Sentence Completion Exercises: Implicit Theories of Leadership and
Leadership Development. In all three conditions, both the Time 1 and Time 2 surveys
ended with a sentence completion exercise to identify the trainee‘s implicit leadership
theories and developmental goal-setting. Two raters (junior leadership instructors with
75
the training institute) naïve to the objective of the study were recruited to rate the
sentences. The raters were asked to provide on a scale of 1 (very poor) to 10 (excellent)
their impressions of the leader with regards to his leadership and his motivation to
improve himself as a leader.
The implicit leadership theories questions in the sentence completion exercise
elicit the trainee‘s assumptions about how leaders ought to behave (e.g. ―To be a leader,
one must be….‖), what constitutes effective leadership (e.g. ―The most effective leader is
one who is….‖), and perceptions of his own leadership (e.g. ―My leadership is
characterized by…..‖). These questions are found in Outcome Measure 1 of Appendix B.
These questions were modified from a series of evaluation discussions that formed part of
the end-of-course ranking exercise for the trainees. The original discussion questions
were ranking questions (e.g. ―Rank the following trainees in descending order in terms of
1) behaviors becoming of an officer 2) effective leadership.) The third sentence
completion question (―My leadership is characterized by …‖) was adapted from a peer
appraisal exercise that required team members to rate each other on a set of desired
organizational values and behaviors. The two raters have prior experience participating in
the end-of-course ranking exercise and peer appraisals from which these sentence
completion tasks were modified from. They were also active duty instructors of junior
leadership courses. As such, they were deemed subject matter experts suitable for
providing evaluations on the quality of these sentence completion tasks.
Evaluations of the each trainee leader‘s goal-setting were made by the two raters
based a goal-setting exercise assigned to the leaders found in Outcome Measure 2 of
Appendix B. The exercise included self-ratings by the leaders on the proportions of time
76
and effort he intended to allocate to self-development, elaborations of his self-
development plans, which included rationale and anticipated obstacles. These rubrics
were extracted from a separate project in the institute to provide another group of trainees
with coaching around an individual development action plan (IDAP). In this project, the
rubrics are provided to coaches as a coaching resource to help coaches more effectively
facilitate goal-setting in trainees. The two raters have been introduced to the coaching
resources. However, this is the first time that the raters were using the rubrics to evaluate
the quality of goal-setting. To familiarize the raters to the rubrics, examples of good and
bad goal-setting statements from the coaching project were provided.
Inter-rater reliability for the ratings provided by the two raters ranged from .83 to
.87. Where ratings differed by more than 2 points on the 10-point scale, the two raters
discussed the difference and adjusted their own scores if required. The final tally
comprised of a single mean score for each of the goal-setting exercise and the implicit
theories sentence completion exercises.
Procedure and Manipulation
At Time 1 (Week 2 of the course), trainees were informed that they will be
participating in 2 surveys regarding the quality of training they have received up to that
point in time. At Time 2 (Week 8 of the course), prior to the survey, trainees were
randomly assigned to one of three conditions – Reflection, Rumination or Control. In
each of these conditions, the trainees were asked to describe two key events from the
training they have received. These two key training events were previously identified
from the qualitative pilot study as formal training activities which provided the highest
number of developmental triggers in the training curriculum. Trainees were then
77
subjected to a 10-minute primer to trigger the various forms of reflection around the two
key training events.
Reflection Manipulation. In the Reflection condition, trainees were asked to
think about how they have benefitted from the training provided by each of the two
formal training activities. They were asked to reflect on how the training can be used in
their future military postings in the unit to train others, as well as how they can adapt the
training to train future trainees in the same course. The aim of these questions was to
prompt trainees to engage in adaptive reflection, specifically getting them to focus on the
utility of the experience and to demonstrate adaptive problem solving. The questions used
to prime the trainees into the abovementioned and subsequent modes of reflection are
included in the Appendix.
Rumination Manipulation. In the Rumination condition, trainees were asked
how the training has inconvenienced them, to detail which aspects of the training most
affected their well-being, and to postulate how the training could have turned out even
worse for them. This train of questioning was designed to get trainees to become self-
absorbed, to engage in extrapolation of negative experiences, both of which are key
ingredients of rumination. In both conditions, the trainees wrote down their thoughts in
their journals. In order to ensure psychological safety, trainees were assured that their
journals were not accessible to the researcher or to the training school.
Control. In the Control condition, trainees were told that the researcher would be
slightly late. They were then assigned a Sudoku puzzle to complete so as to ‗pass the
time‘. To ensure task motivation, trainees in this condition were told that the one to
complete the puzzle the fastest will be given extra time for a break at the canteen at the
78
end of the day. This Sudoku puzzle required an average of 10 minutes to complete,
thereby equalizing the cognitive load of trainees in the Control condition compared to
those in the other two conditions who had to engage in the journaling activity. Once the
primers were completed, the Time 2 survey was administered to the trainees.
Manipulation Checks
A three-item manipulation check was administered with the trainees from both the
basic military training course and the leadership training course at the end of the primer
in each condition at Time 2.
Reflection/Rumination invokes emotional processing. The first item evaluated
the extent to which reflection or rumination about the key training events invoked
emotional processing that differed from other forms of pure intellectual activities such as
solving Sudoku puzzles such as those experienced by the participants in the control
condition. (How did you feel about the event/activity you were told to think about?
Anchors were 1=very negative to 5=very positive). Because the events to be recalled and
processed were significant milestone events, there should be more emotional reactance
demonstrated by participants who were either reflecting or ruminating, compared to those
in the control treatment in which cognitive resources were invested in a distraction
exercise (Sudoku puzzle). As such, it is expected that any form of processing of the
recalled events should result in more emotional reactance compared to the control, with
rumination demonstrating more negative reactance than reflection. The means and s.d. for
the manipulation check item 1 analysis is found in Table 6.
------------------------------------------------------------
Insert Table 6 Here
79
------------------------------------------------------------
A one-way ANOVA revealed a significant difference across the three groups for
both the low leadership experience condition, F(2, 457)=8.00, p<.001 and the high
leadership experience condition, F(2, 328)=4.77, p=.009.
In the low leadership experience condition, planned contrasts revealed that
compared to the control, there was a significant difference in the emotional reactance
experienced by the trainees who underwent any form of reflection or rumination,
t(457)=3.22, p<.05, with the reflection/rumination groups receiving more negative
reactance than the control as expected. Likewise, there was also a significant difference
in the emotional reactance experienced by the trainees in the rumination versus reflection
group t(457)= 2.27, p<.05, with the rumination group being more negative in the
reactance than the reflection group as expected.
In the high leadership experience condition, the same pattern repeats itself in the
planned contrasts. There was a significant difference in the emotional reactance
experienced by trainees who underwent any form of reflection or rumination,
t(328)=3.08, p<.05, with the reflection/rumination groups receiving more negative
reactance than the control as expected. However, this time round, though the rumination
group demonstrated more negative reactance than the reflection group, the difference was
not significant t(328)=.67, n.s, possibly due to the smaller sample sizes in both the
reflection and rumination treatment groups.
Regardless, in both low and high leadership experience samples, it was
demonstrated that any form of reflecting about the events led to some form of emotional
reactance. This reactance is indicative of emotional processing, which meant that the
80
reflection/rumination manipulations are indeed invoking processing different from the
pure cognitive tasks experienced by the control groups in both high and low leadership
experience samples.
Rumination intervention is valid. Rumination is associated with clinical
depression (Nolen-Hoeksema, 2000). The second manipulation check item leverages this
effect to determine whether the rumination primer did indeed create a differential effect
from the reflection and control primers. The anchors for this item was 1=Very Depressed,
3=Depressed and 5=Not depressed at all. The means and standard deviations for
manipulation check item 2 can be found in Table 7.
------------------------------------------------------------
Insert Table 7 Here
------------------------------------------------------------
Once again, a one-way ANOVA revealed a significant difference between the
groups for the low leadership experience condition, F(2,456)=6.32, p=.002, and for the
high leadership experience condition, F(2,328)=6.35, p=.002.
In the low leadership experience condition, planned contrasts revealed that
rumination compared to any other form of cognitive activity (i.e. comparing rumination
with reflection and control) result in a significantly higher level of reported depressive
feelings, as expected, t(457)=2.67, p<.05. This effect is also replicated in the high
leadership experience condition t(328)=2.88, p<.05. This lends support demonstrating
that the rumination intervention is indeed valid, as it demonstrates depressive self-reports
as predicted by literature.
81
Reflection intervention is valid. Reflection is associated with critical thinking
and problem-solving, with a focus on positive aspects of an experience, openness and
learning from experience (Dewey, 1933). The first manipulation check question
examined whether reflection of any kind triggers greater processing of an experience.
The second manipulation check question confirms that the direction of the rumination
trigger is indeed demonstrating one aspect of rumination which is self-reported
depressive feelings. The third and final manipulation is a check on the more distal effect
of the reflection manipulation, which is to replicate the effect of higher self-reported
feelings of hope that should arise from having engaged in problem-solving modes of
thinking compared more ruminative ones. The anchors for this item was 1=Very Hopeful,
3=Hopeful, and 5=Not Hopeful at All. The means and standard deviations for
manipulation check item 3 can be found in Table 8.
------------------------------------------------------------
Insert Table 8 Here
------------------------------------------------------------
A one-way ANOVA revealed a significant difference between treatment groups
for the low leadership experience condition, F(2,451)=13.41, p<.001, and for the high
leadership experience condition, F(2,328)=11.69, p<.001.
In the low leadership experience condition, planned contrasts between reflection
against rumination treatment groups revealed a significant difference in self-reported
hope, with trainees in the reflection condition reporting higher levels of expressed hope
about the events as compared to the rumination condition , t(451)=3.18, p<.05. In the
high leadership experience condition, there is no significant difference, possible due to
82
the reduced cell sample sizes t(328)=1.48, n.s. Hence, it appeared that while the reflection
manipulation was powerful enough to elicit processing of the events in the low leadership
experience condition, the effect was not powerful enough to be significantly
demonstrated in a reduced cell sample size in the high leadership experience condition.
In summary, the reflection and rumination primers did result in greater emotional
(albeit more negative) reactance and higher levels of hope than those in the control. Of
the two, the rumination primer also elicited higher levels of self-reported feelings of
depression than the reflection primer as expected from literature. This demonstrates the
rumination primer to be successful in terms of magnitude of effect created and direction
of the effect. The reflection primer performed similarly to the rumination condition in
eliciting higher levels of processing of events. However, it was not sufficiently powerful
enough to allow the trainees to positively reframe the negative events to engender higher
levels of self-reported hope.
Establishing initial equivalence. In order to establish initial equivalence, in both
the high and low experience samples, trainees at Time 2 were randomly assigned to one
of the three conditions. To verify that initial equivalence was established, trainees‘
quality of reflection outcomes, trait reflectiveness, Psycap, GLE, and general self-
efficacy were compared at Time 1. This was verified, with all the scales not significantly
different across the three conditions.
83
CHAPTER 4: RESULTS
84
Chapter 4: Results
As a prelude to discussing the specific data analyses and results of hypotheses
testing, the workflow for data screening, including dealing with missing data, outliers,
and normality is first presented. Following the preliminary analyses, the main analyses
and results for each hypothesis are presented. Analyses include correlation analysis, t-
tests, ANOVAs, MANOVAs and structural equations modeling. Finally, this chapter
concludes with supplemental analysis and a summary of the results.
Data Screening
Before conducting hypothesis testing, the data was entered, cleaned, and prepared
for analysis in SPSS. All reverse-scored items were then recoded. Next, an outlier
analysis was conducted for items.
Outlier Analysis
Because multivariate analyses can be very sensitive to outliers, or data points that
are very different from the rest of the data set, preliminary analyses were conducted to
examine the extent of outliers present in the data. Outliers due to data recording or data
entry error were eliminated by the use of the online survey system as noted in the
previous chapter. However, outliers due to participants responding very differently from
the rest still needed to be accounted for. This was particularly important because the
online survey system did not allow the respondent to proceed until all entries for any
given page was completed, which meant that an unintended consequence could be that
the unmotivated respondent may be inclined to provide patterned responses (e.g. enter in
85
all 1s or all 7s) just to complete the survey. Such patterned responses will also contribute
to outliers in the dataset.
To address the issue of outliers, a common rule of thumb of univariate outliers
was used where values more than three standard deviations away from the mean were
considered an outlier. Through the examination of the frequency distribution of the z-
scores for all study variables, the initial extent of univariate outliers were identified.
Distortions to multivariate relations due to outliers can be reduced when items are a part
of scales. Hence, it was also prudent to check for the extent of outliers at the scale level.
At the scale level, a total of 62 of 5,163 data points had z-scores larger than three and
consequently 1.2% of the data points were considered outliers. At the item level, 1,074
out of 81,134 data points or 1.32% were outliers.
To correct for outliers, it must first be understood why the outliers occurred. For
example, the issue could have been due to attributes of the items, such as their
interpretability. On the other hand, outlier responding could also have occurred as an
outcome arising from some characteristics of the respondents.
To examine for respondent characteristics associated with outliers, a physical
inspection of the dataset itself was conducted to scan for patterned responding across all
items. At Time 1, 7 such cases were identified (cases no 123, 164, 165, 195, 233, 267 and
294). At Time 2, 3 cases were identified (cases no 13, 220 and 258). I interpreted these
responses as having been induced by lack of motivation at that point in time. Hence, I
decided to remove these responses from the dataset and treated these as missing values.
Next, outliers at the scale level that fell 3 standard deviations away from the mean
were examined. Again, the focus here was on trying to understand the pattern of response
86
demonstrated in the entries of all the items in each scale. Several cases stood out as
potential unmotivated responding. Cases 168 and 185 scored all 5s and 9s across the
quality of reflection outcome scales and the GLE scales at Time 1, while Case 95
demonstrated a similar pattern across the above scales as well as for Psycap. Their entries
were deleted and treated as missing. In addition, Case 185 also displayed similar extreme
values across the Psycap and Trait reflection scales. Given the propensity for Case 185 to
demonstrate extreme responses, its entire Time 1 entries were also deleted.
Aside from these, there were other cases that displayed extreme scores at the scale
level. Case 294 entered all 5s for the GLE scale, but the rest of the response appeared
normal. For instances like these, the scores for those particular items were deleted so that
they could be estimated subsequently using maximum likelihood procedures.
Finally, there were cases which the patterned responding occurred close to the end
of the survey. Case 109 appeared to have normal responding until about the final third of
the survey, where his responses were all 1s. I interpreted these as ill-considered responses
driven by a lack of time and deleted only these entries.
Taking the above actions to address outliers, the issue of univariate outliers was
addressed.
Normality
To assess multivariate normality, skewness and kurtosis statistics were examined
for each variable. While a normal curve has skewness and kurtosis values of zero, values
between +/- two and +/- seven for skewness and kurtosis, respectively, are considered
acceptable (Kline, 1998). None of the scale composites from the self-report data had
skew or kurtosis values which exceeded these rules of thumb.
87
------------------------------------------------------------
Insert Tables 9-12 Here
------------------------------------------------------------
Initial equivalence testing
To ensure that there was initial equivalence across experimental conditions,
participants from the study sample were randomly assigned to one of the three
manipulations. To verify that initial equivalence was indeed achieved, a MANOVA of all
variables at Time 1 was conducted. None of the variables were significantly different
across treatment groups, although trait reflection came closest, F(2,318)=2.60, p.=.076.
Given this finding and given the fact that trait reflection could influence the effort and
quality of reflection across treatment groups, it was included as a covariate in subsequent
analyses and modeling.
Hypotheses Testing
Figure 3 depicts the structural relationship between Psycap, GLE and Trait
Reflection taken from the high leadership experience sample at Time 1 and the Quality of
Reflection outcomes taken at Time 2. It also captures the beta weights of the subscales to
the second-order latent factors of Psycap, GLE and Quality of Reflection outcomes.
Though the CFI was lower than the cut-off of .950 used in the pilot studies, the overall fit
of the model was acceptable (SRMR=.056, CFI=.904, RMSEA=.036) and still within the
guidelines suggested by Bollen‘s combinatory rule of SRMR < .08 and either RMSEA
<.06 or CFI >.95.
88
------------------------------------------------------------
Insert Figure 3 Here
------------------------------------------------------------
The beta weights of the Time 1 latent factors on Quality of Reflection outcomes at
Time 2 are all not significant. This again lends support for the success of the
manipulation, which is that Quality of Reflection is trigger-dependent rather than state-
dependent, i.e. due more to the nature of the reflection triggers than to the characteristics
of the person.
Hypotheses 1a-1d
H1a: The level of Psycap is positively correlated with prior leadership
experience.
H1b: The level of leadership efficacy is positively correlated with prior leadership
experience.
H1c: The level of leadership self-awareness is positively correlated with prior
leadership experience.
H1d: The quality of implicit leadership theories is positively correlated with prior
leadership experience.
What is the effect of leadership experience on the level of Psycap, leadership
efficacy, leadership self-awareness and quality of implicit leadership theories? Firstly, an
examination of the simple means (refer to Tables 9 & 10) indicate that the differences in
means between the two levels of leadership experience are generally in the expected
directions. To test for H1a-d, a ONE-WAY ANOVA was conducted to explore the effect
of leadership experience on the outcomes of Psycap, GLE, leadership self-awareness and
89
quality of implicit theories at Time 2 using both the low and high leadership experience
samples combined. Of these, Psycap was not significantly different between the high and
low leadership groups F(1,826)=2.11, p=.147. Hence, H1a is not supported.
Leadership efficacy was significantly higher in the sample with higher leadership
experience, as predicted by H1b, F(1, 826)=14.40, p<.001. Leadership self-awareness did
not differ significantly between the two groups, F(1,826)=1.03, p=.748. Hence, H1c is
not supported. Finally, the quality of implicit theories was also significantly higher in the
group with the higher level of prior leadership experience F(1, 713)=7.76, p=.005. Hence,
H1d is supported.
Hypotheses 2a-c
H2a: The relationship between leadership efficacy and leadership experience is
moderated by type of reflection triggers.
H2b: The relationship between leadership self-awareness and leadership
experience is moderated by type of reflection triggers.
H2c: The relationship between quality of implicit leadership theories and
leadership experience is moderated by type of reflection triggers.
To test for the presence of a moderation effect of rumination vs reflection triggers
on the impact of prior leadership experience on leadership efficacy, leadership self-
awareness and quality of implicit leadership theories, 3 (reflection, rumination, control) x
2 (low vs high leadership experience) MANCOVA with trait reflection and trait self-
efficacy as controls was conducted on the combined low and high leadership experience
samples using the data captured at Time 2, and using control variables captured at Time
1. The interaction terms between the predictor (leadership experience) and the moderator
90
(type of reflection) was examined, with significant interaction terms indicating the
presence of moderation.
Interpreting the MANCOVA results. For the MANCOVA to be appropriately
applied there are three assumptions that need to be met. Firstly, the MANCOVA assumes
that the errors are independent of each other across observations and the independent
variables in the model. This assumption is valid as the grouping variable of leadership
experience was from different samples while the other grouping variable of reflection
type was randomly assigned to trainees in both leadership experience levels.
Secondly, MANCOVA also assumes that covariances of dependent variables are
constant across cells. This assumption is critical when cell sizes are not uniformly the
same, as was the case in this study. The issue is not problematic for the sample with
lower leadership experience. However, it was a greater problem for the sample with
higher leadership experience. To ascertain the extent of this problem, the MANCOVA
was first conducted with the full sample and then repeated with randomly selected
trainees from each cell such that cell sizes are equal. The results are similar in pattern.
Hence, the full sample analysis will be reported here.
The final assumption of the MANCOVA is that of multivariate normality. This is
typically difficult to ensure in practice. Nonetheless, univariate normality was examined
and found to be within acceptable ranges.
Given that some of the assumptions of the model may be violated, Pillai‘s trace
will be the reference statistic as it is more robust to such violations (Olsen, 1974). Pillai's
trace is a positive-valued statistic, with higher values indicating effects that contribute
more to the model. Pillai‘s trace will also be compared against the Hotelling's trace,
91
which is the sum of the eigenvalues of the test matrix. It is a positive-valued statistic for
which increasing values indicate effects that contribute more to the model. When both
Pillai‘s and Hotelling‘s traces are small and comparable, it indicates that the effect does
not contribute much to the model. Should this happen, further triangulation is derived by
looking at the partial eta squared statistic, which reports the "practical" significance of
each term, based upon the "ratio" of the variation accounted for by the effect to the sum
of the variation accounted for by the effect and the variation left to error.
Given the above background information, the MANCOVA results indicated that
the type of reflection by level of prior leadership experience interaction term was
significant F(6,1540) = 2.66, p=.014, η2 = .01. The interaction was significant for
leadership efficacy, F(2,779)=3.97, p=.02, η2 = .01 and for leadership self-awareness
F(2,779)=4.20, p=.02, η2 = .01, but not for quality of implicit leadership theories
F(2,779)=2.60, p=.08, η2 = .01. Hence, H2a & b were supported but not H2c.
------------------------------------------------------------
Insert Figure 4-6 Here
------------------------------------------------------------
Post-hoc Explorations & Hypotheses 3-5
In the literature review, the possibility of an interaction of leadership experience
and type of reflection triggers was speculated to influence leadership efficacy, leadership
self-awareness and quality of implicit leadership theories either positively or negatively.
Given this, the effect of reflection triggers was inspected separately post-hoc under
conditions of low and high leadership experience. To do this, effects of type of reflection
triggers under low and high leadership experience conditions were examined by
92
employing a ONE-WAY ANOVA with post-hoc comparisons on each low and high
experience sample separately. Hypotheses 3 to 5 will also be examined separately under
conditions of low and high leadership.
H3a: Recalling and reflecting positively on any personal leadership experience
increases subsequent leadership efficacy.
H3b: Recalling and reflecting negatively on any personal leadership experience
decreases subsequent leadership efficacy.
H4a: Recalling and reflecting on past experience results in greater self-
awareness.
H4b: Ruminative recall of past experience results in greater self-awareness
compared with adaptive recall.
H5a: Adaptive reflection results in positive aspects of one’s implicit leadership
theories becoming more easily explicated.
H5b: Maladaptive reflection results in negative aspects of one’s implicit
leadership theories becoming more easily explicated.
Low leadership experience. Under the low leadership experience condition, the
pattern of mean scores across type of reflection triggers were consistent with
expectations, which was that one would expect reflection triggers to score highest,
followed by control and then rumination. This was true for all the outcomes (see Figures
4-6). In other words, the simple effects of type of reflection triggers on the outcomes
were occurring in the expected direction. However, not all the cell mean differences or
simple effects were significant. The post-hoc comparisons with Bonferonni correction
reveal no significant difference under reflection versus rumination trigger conditions for
93
leadership efficacy, t(494)=1.29, p=.20 and leadership self-awareness t(494)=.816, p=.42.
Only the quality of implicit leadership theories differed significantly with it being higher
under reflection versus rumination triggers t(442)=4.28, p<.001. Hence, under conditions
of low leadership experience, the only support found was for H5a & b.
High leadership experience. Under the high leadership experience condition, the
pattern of mean scores across type of reflection triggers was again consistent for quality
of implicit leadership theories, with it being higher in reflection versus rumination
triggers t(328)=2.14, p=.033. Hence, again, support for H5a & b was found.
On the other hand, contrary to expectations, an inspection of the mean scores
revealed that it was the rumination rather than the reflection triggers that achieved
significantly higher levels of leadership self-awareness t(328)=-2.86, p=.005 and
leadership efficacy t(328)=-2.53, p=.012. Hence, while there are significant differences
across type of reflective trigger conditions, there is no support for H3a & b regarding
effect of type of reflection on leadership efficacy and for H4a & b regarding effect of
type of reflection on leadership self-awareness.
Predictors of GLE at Time 2 for high leadership experience. In addition to
testing the hypotheses, it appeared from the qualitative interview data that there was
something distinctly different between the responses of the ‗better‘ leaders versus the
others. Realizing that my own mental models or implicit theories of leadership may be at
play in segregating these ‗better‘ leaders from the others, I determined that my own
implicit theories of leadership contained a huge element of high leadership efficacy for
those who came across as good leaders in the interviews. Hence, I decided to explore
what other variables may be associated with GLE.
94
From the datasets available to me to conduct this exploratory analysis, the
predictors of GLE at Time 2 were modeled and depicted in Figure 7 below. The model fit
was adequate given the complexity of the model (SRMR=.055, CFI=.900,
RMSEA=.046). The beta-weights for the paths from Psycap and leadership self-
awareness at Time 2 to GLE at Time 2 were both significant at the 95% confidence
interval (β=.928 and .790 respectively).
An alternative model was tested in which Psycap mediates the relationship
between leadership self-awareness and GLE at Time 2. This represents a more restrictive
model on which a chi-sq difference test can be applied. Should the test turn out non-
significant, then there is no difference between both models and so the more
parsimonious (ie restrictive) model should then be accepted. In this case, the chi-sq
difference test was highly significant (chi-sq diff = 7.162, d.f. diff = 1, p<.001). Hence,
the original model of Psycap and leadership self-awareness being predictors of GLE at
Time 2 was accepted instead.
------------------------------------------------------------
Insert Figure 7 Here
------------------------------------------------------------
Using the above comparison methods, I tested several other models using Time 1
predictors against GLE at Time 2. However, these either turned out to either provide
much poorer fit or simply failed to converge properly. Hence, it was safe to exclude Time
1 predictors from the final model. Hence, my implicit theories of good leadership,
evidenced by high GLE at Time 2, were statistically connected to the level of Psycap and
leadership self-awareness. In other words, my experience of encountering ‗good‘
95
leadership during my interviews are likely due to these leaders having a high level of
GLE, Psycap, and whose leadership self-awareness are high.
Statistically, the final model was also equivalent to another alternative model in
which the leadership self-awareness at Time 2 was the criterion, and in which GLE was a
mediator. This model, though theoretically feasible, was too much of an inference to be
drawn from a dataset in which both predictors and criterion were not temporally
differentiated.
96
CHAPTER 5: DISCUSSION
97
Chapter 5: Discussion
Review of Purpose of Dissertation
This dissertation explored triggering reflective thinking and its impact on
variables important for leadership development, such as leadership efficacy, leadership
self-awareness, one‘s implicit leadership theories of leadership and one‘s psychological
capital. Specifically, this dissertation sought to establish whether it was possible to create
situational triggers that impacted different types of reflective thinking, and whether these
types of micro-interventions were significantly powerful to impact the leadership
development variables.
Overview of Discussion of Results
The manipulation checks in the pilot and actual samples confirmed that the
triggers successfully replicated the range of ruminative to reflective effects intended for
in the dissertation. In both the low and high leadership experience samples, both the
reflection and rumination triggers worked as intended, according to the manipulation
check items. The relevant literature supporting the choice and outcomes of the
manipulation items have been provided in Chapter 3 and will not be repeated here.
In actual use, however, the reflective triggers appeared to have a much weaker
effect on leadership efficacy, leadership self-awareness and than that of the ruminative
98
effects in the sample with higher leadership experience. These results will be discussed as
part of the post-hoc analyses.
On the other hand, the reflective effects were stronger than the ruminative effects
in the sample with lower leadership experience, suggesting the presence of an interaction
between leadership experience and the type of triggers. Though speculated, the presence
of a significant interaction effect (Figures 4-6) was not initially part of the literature
review and will therefore be discussed in detail.
There are several plausible explanations for leadership experience to interact with
the effect of type of triggers. The first type of explanation is along the lines of threats to
validity, and will be elaborated when the limitations of the study are examined. The
second type of explanation is theoretical in nature and will be elaborated in the section on
theoretical contributions.
Effect of Leadership Experience: Discussion of Hypotheses 1a-d
Hypotheses 1a-d explored the correlation of leadership experience with Psycap,
leadership efficacy, leadership self-awareness and quality of implicit leadership theories.
The analyses revealed that in line with H1b & d, leadership efficacy and quality of
implicit leadership theories were positively correlated with leadership experience. The
rationale for these has been discussed in Chapter 1 and will not be repeated here. Rather,
Psycap, which was also influenced in the expected direction but not significant, and
leadership self-awareness, which was not influenced by leadership experience, will be
discussed in greater detail.
Lack of effect on Psycap. There was an improvement in Psycap with greater
leadership experience, except that this was not significant. A possible reason could be
99
due to the choice of samples used in this dissertation. The low leadership experience and
the high leadership experience samples differed from each other by only nine weeks.
Although Psycap has been demonstrated to be improved in micro-interventions (Luthans,
Avey, Avolio, Norman, & Combs, 2006), this was not the case here because the training
course was not designed to implement Psycap improvement. Nonetheless, because of the
relevance of Psycap to leadership development, it is not surprising that here Psycap is
improving with more military leadership training albeit not significantly so.
Lack of effect on leadership self awareness. Leadership self-awareness was not
influenced by leadership experience. This was not a surprising finding on hindsight.
While one may have predicted that more experienced leaders should have demonstrated
more leadership self-awareness, again the choice of samples used in this dissertation,
with both samples differing by only nine weeks in terms of soldiering and military
leadership, would have been too close together to expect any significant differences in
leadership self-awareness.
Viewed in another way, this finding indirectly provides support that the state of
leadership self-awareness is more amenable to situational triggers. Further support for
this is found in the structural equations modeling (refer to Figure 3) in which none of the
other psychological variables such as Psycap, leadership efficacy and trait reflection.
Interaction of Leadership Experience and Type of Reflection Triggers: Discussion of
Hypotheses 2 to 5
There is a significant interaction between the level of leadership experience and
the type of reflection triggers. This interaction was significant for leadership efficacy
(H2a) and leadership self-awareness (H2b), and only marginally non-significant for
100
quality of implicit leadership theories (H2c). These results will be discussed under
conditions of low and high leadership experience. (Refer to Figures 4-6).
Implicit leadership theories. Reflection triggers produced significantly higher
levels of implicit leadership theories under both low and high leadership experience
conditions (i.e. H5a & b were supported). The benefit of using reflection triggers is most
marked under conditions of low leadership experience, with reflection triggers
performing significantly better than rumination triggers. However, with greater
leadership experience, the differential effects of types of triggers were no longer as
apparent (i.e. H2c marginally unsupported).
The above findings are not surprising. As described in the literature review,
implicit leadership theories are knowledge structures. When one is naïve about
leadership, one‘s ideas and assumptions about leadership are less well-formed and hence
more amenable to external influence. Reflection triggers, being designed to be thought-
provoking, is therefore more likely to have a stronger influence on the expression of ideas
of leadership. When leadership experience is high, there is more available stable self-
knowledge about oneself as a leader. Hence, rumination, with its self-focused attention,
should therefore not fare too badly compared with reflection, as was found here.
Leadership efficacy. In both the low and high leadership experience samples,
support for H3a & b were not found. In the low leadership experience condition,
reflection triggers resulted in higher levels of leadership efficacy but was not significant.
On the other hand, in the high leadership experience condition, the reflection triggers
performed significantly worse than the rumination triggers. This was a surprising finding.
101
In the literature review in Chapter 1, I proposed that rumination is counter-
productive for concurrent problem-solving because it channels finite cognitive resources
from problem-solving to attend to the self. I then predicted that reflection with its
problem-focused approach should fare better. In the long run, leaders who reflect rather
than ruminate will engage in more problem-solving thinking, which should result in
increased leadership efficacy. In this instance however, this train of argument was not
sufficiently supported by the data from the low leadership experience group and contrary
to the data from the high leadership experience group. Regarding the former, it may be
that the reflection triggers were not repeated frequently enough throughout the training
course to generate enough leadership efficacies through repeated enactive mastery
(Bandura, 1997). This was not the intent, as the dissertation aimed to explore the newer
approach to raise leadership efficacy through a micro-intervention of varying the type of
reflection trigger, rather than through the already established route of enactive mastery
through repeated practice.
Regarding the latter finding of rumination performing better than reflection in the
high leadership experience condition, firstly, it must be noted that in spite of this finding,
the aim of designing a micro-intervention for triggering higher levels of leadership
efficacy is still met, except that the novel finding here is that rumination works better
than reflection when leadership experience is high.
Next, there are several possible explanations for why rumination should perform
better than reflection in the high leadership condition. First, it was uncovered through the
pilot study with this group of trainees that the training environment in the high leadership
experience condition was highly arduous and negative. Under such conditions, the focus
102
of reflection to negative aspects may actually facilitate catharsis and recovery, thereby
helping the trainees to deal better with the training and hence achieving enactive mastery.
Second, familiarity with the ruminative over the reflective process may be another
reason. The pilot study uncovered that the mode of after-action reviews practiced by the
trainees tended to focus on the negative aspects (―what went wrong‖) of the experience.
The training curriculum in the high leadership experience condition was very intensive
and not facilitative of the more novel reflective habits from being assimilated and
practiced. Hence, when reflection triggers were introduced during the experiment at the
end of the course rather than habituated each time dilemmas occur, trainees are less likely
to adapt well as they are less familiar with this form of reflection. Given these
background reasons, it is plausible for trainees exposed to the ruminative trigger
condition to outperform trainees in the reflection condition.
Leadership self-awareness. The pattern of findings for leadership self-awareness
is similar to that for leadership efficacy. In both the low and high leadership experience
samples, support for H4a & b were not found. In the low leadership experience condition,
reflection triggers resulted in higher levels of leadership self-awareness but was not
significant. On the other hand, in the high leadership experience condition, the reflection
triggers performed significantly worse than the rumination triggers for leadership self-
awareness, as was the case for leadership efficacy.
As with the situation for leadership efficacy, familiarity with the rumination
process in the high leadership experience may have resulted in the greater potency of the
ruminative trigger over the reflection trigger. On the other hand, the focus of the training
course may also have played a role.
103
In the high leadership experience condition, junior leader trainees were trying to
acquire the knowledge, skills, abilities and competencies of military leadership. In fact,
the training objectives of the course were explicit in the leadership and vocational skills
needed, but less so on the leader identity, values and purpose. Hence, it is logical to
assume that the focus of after action reviews during training therefore also on skill
development, with less time available to touch on the deeper leader identity-development
aspects that the scale attempted to capture. Hence, it is possible that with less time and
attention on identity development, there was less experienced change for the reflective
triggers to activate during the end-of-course intervention.
Post-Hoc Analyses: Predictors of leadership self-awareness. Using the datasets
firstly from the low leadership experience cohort and then repeated on the high leadership
experience cohort, the full structural model was constructed to determine the extent of
contribution of psychological constructs such as Psycap and leadership self-awareness
toward leadership efficacy (Refer to Figure 7 for final structural model on high leadership
experience cohort). The intent of the exploratory analysis was to understand the relative
contribution of the new leadership self-awareness construct on leadership efficacy, given
the contribution of Psycap. It was heartening to note that firstly, Psycap was uncorrelated
with leadership self-awareness, providing assurance of the construct validity of newly
constructed leadership self-awareness scale. Secondly, the relative contribution of
leadership self-awareness on leadership efficacy was very high (β=.79) even after taking
into account the presence of Psycap. However, more research is needed to refine the
reliability and validity of the newly constructed scale before further inferences can be
made from this full structural model. Nonetheless, at least for this sample, raising the
104
leadership self-awareness of trainees, which has been demonstrated to be feasible through
the use of reflective triggers, should significantly raise the leadership efficacy of trainees,
thereby contributing to the leadership development of these trainees.
Potential Limitations
This dissertation, like most, has limitations that may potentially restrict its internal
and external validity. Internal validity refers to the degree in which cause-effect
relationships postulated in the study are true. When internal validity is high,
experimentally manipulated effects occur as intended by the researcher and are not due to
other causes. In other words, observed effects cannot be reasonably explained by other
factors. These other factors are also known as threats to validity. The likelihood that such
factors threatened the validity of the dissertation should be examined. In this dissertation,
the potential threats to internal validity are: climate, effectiveness of the reflection-
rumination manipulations, participants‘ motivation, uneven sample distribution and
power, limitations of measurements and method and finally, experimenter bias.
External validity refers to the extent to which results concluded from a study are
generalizable to other contexts or population groups. When external validity is high, it is
typically because the context or samples employed in the study are highly comparable to
other more general contexts or population groups. In this dissertation, the potential threats
to external validity are the training context and characteristics of the sample itself.
All threats to validity will be discussed together in the following section.
Climate. The single biggest limitation of the dissertation came from the
unexpectedly negative training climate uncovered. This negativity manifested itself in
how trainees were conditioned to think. Examples of the artifacts of the negative climate
105
included increased performance-avoid orientations after undergoing training, focus on
negative aspects of performance during AARs, high levels of cynicism and perception of
double standards in instructors.
Effectiveness of the reflection-rumination manipulations. The reflection
manipulation was not as powerful as the rumination manipulation. Worse, the reflection
manipulation likely interacted with the negative climate to influence the efficacy
outcomes in the opposite direction.
Participants’ motivation. Participants may have had varying levels of motivation
to complete the intervention. The training curriculum, as previously mentioned, was very
intense. All the surveys and interventions occurred during limited free periods, which
typically tended to be just before dinner during evenings. With limited free time,
participants may have been distracted and may have found it hard to settle down to a
balanced state to respond to the triggers as intended. As the sessions were conducted
prior to dinner, they may also have been motivated to complete the study quickly, thereby
demonstrating a superficial level of effort that is not congruent with that required by the
study, particularly the reflection condition.
Another aspect of participant motivation could have originated from familiarity.
The after-action review steps in the rumination condition are more familiar to the trainees
than the ones in the reflection condition. This could have translated into higher
performance for the rumination condition while the lack of practice in the reflection
condition could have suppressed performance.
Given the above, my assessment of the threat to validity due to participants‘
motivation is high. This therefore implies that there could have been a general
106
suppression of performance due to potential distraction encountered by participants.
Between reflection and rumination, participants could also have under-performed in the
reflection condition due to unfamiliarity.
Uneven Sample Distribution and Power. A third limitation concerns the uneven
distribution of subjects to the rumination and control conditions. I was constrained by
training requirements to allocate more subjects to the condition that was expected to
receive more benefit. Hence, I allocated extra trainees to the reflection condition. In
hindsight, doing so has led to reduced power in the rumination and control conditions.
Measurement Limitation. Another limitation relates to the use of new measures
as outcome variables. The GLE and Quality of Reflection Outcomes were new measures.
While the Psycap scale has more established validity research, it was also relatively new
as well. In of themselves, the use of new measures is not a problem if these measures are
reliable and valid. In this study, the reliabilities and validities of the measures were
acceptable. The issue with the use of new measures is the risk of capitalizing on chance
when there is not enough evidence for the robustness of the measures when applied to a
different culture, context or sample. This was definitely the case here as all the measures
were applied to an entirely new culture and sample.
Method Limitations. Next, all the data collected were from survey-based
methods or from focus groups. Although objective data, particularly performance data
were initially planned for in the dissertation proposal, this was eventually unavailable due
to national security issues. Ideally, multiple sources of data should have been ensured to
reduce the risk of introducing a method bias to the study.
107
Experimenter Bias. Experimenter bias occurs when participants in a study are
influenced by the cues or expectations projected (consciously or otherwise) by the
experimenter, or when the experimenter‘s judgment is influenced by his/her own
expectations. In this dissertation, the areas whereby this could have occurred are 1)
during survey administration 2) during coding of the qualitative data. To reduce the risk
of experimenter bias, survey administration was conducted by assistants who were naïve
to the design and intent of the study. During survey administration, they adhered to a
script. Secondly, the coding of the qualitative data was conducted by assistants who were
similarly naïve to the design and intent of the study. During the coding, the assistants
were assigned random piles from all three treatment conditions to score. Despite this, the
inter-rater reliability reported remained high. Hence, on the whole, experimenter bias was
not assessed to be a significant threat to validity.
Sample. The sample used in the study may potentially limit the generalizability of
the findings. For example, the subjects were all male emerging adults with high military
leadership potential training in a military environment. It remains to be seen if females
would manifest different coping and performance in such a similar environment. On the
other hand, because the subjects were drawn from a national compulsory conscription
cohort, they form a good representation of all males with leadership potential in the
country, taking national demographics of race, religion, education into account. Hence,
while the sample does not include females, the findings in this dissertation should be
applicable to emerging adult males with leadership potential in this country.
Training Context. The training context in this dissertation was unique in several
ways. Firstly, the training environment was applied to a captive audience. Being
108
conscripted, trainees in the study had varying levels of motivation to want to lead or to be
trained as leaders. This directly impacts the quality of reflection that may be induced in
the study.
Secondly, the training context was military in nature. The pervasiveness of the
military milieu, such as regimentation, high power-distance between trainees and
instructors the military focus, content and application of training presented unique
challenges for me to introduce reflective practices. Compared to the military, reflective
practices are more prevalent and established in other contexts such as education or
business. Hence, reflective practices may come more naturally to these contexts than in
the military. This will again directly impact the quality of reflection that may be induced
in the study. On the positive side, however, the relationships identified in this study may
therefore be a conservative estimate of that which may be uncovered in more natural
contexts such as education and business.
Recommendations and Implications for Future Research
Developmental climate. Future research should include developmental climate as
a variable to be examined further. In this dissertation, it appeared that the operating
climate played a role in influencing the impact of the reflective triggers. In a quasi-field
experiment, Ellis and Davidi (2005) demonstrated that when after action reviews focused
only on failed experiences, soldier performance in field navigation were poorer then
when the after action reviews focused on successful experiences as well. The authors
attributed the better performance to the inclusion of both successful and failed
experiences as part of the after action reviews. However, another plausible explanation
may be that the inclusion of both types of experiences may be perceived as a more
109
developmental climate in play, i.e. one that encourages after action reviews and
reflections to occur, and in which emphasis is placed on learning from both positive and
negative events.
In this dissertation, the lack of such an emphasis may have reversed the effects of
ruminative and reflective triggers on leadership development. So what might constitute a
positive or negative developmental climate? In perhaps a more normative climate,
trainees engaging in rumination should have performed poorer than trainees engaging in
reflection in leveraging on the developmental potential of triggers encountered. Trainees
who engage in rumination invest their limited cognitive resources on unproductive
thinking such as dwelling on symptoms rather than on solutions. Conversely, trainees that
engaged in reflection that is future oriented, problem focused should have derived new
insights to problems faced, thereby increasing their leadership efficacy and Psycap.
In a positive climate, the advantages of reflection over rumination should be even
more significantly felt. In such a climate, it should be expected that the climate would
predispose reflective practices and reflective thought. With institutional investment in
resources and time for reflection, those who whittle away their opportunities for
reflection and engage in rumination instead should find themselves further afield than
those who have been compliant to the climate.
In an oppressive climate, trainees may actually be experiencing leadership-related
negativity or even trauma. Examples elicited in the pilot study included trainees
observing instances of leaders who failed to walk-the-talk, demonstrated double
standards, were poor role-models, and who were disinterested in trainee development. It
is outside the scope of this study to understand how such leadership behaviors can be
110
found in a training institute, though I speculate that instructor burnout may play a large
part in it. Nonetheless, when trainees are demanded to demonstrate high standards
expected of leaders and yet do not see the same standards in their instructors, this
discrepancy may lead to a negative reaction when asked to display any form of leadership
they do not see role-modeled in the instructors.
Competing demands. Cognitive resources are needed for self-development.
Likewise, cognitive resources are also required to maintain harmony within one‘s internal
psyche. In the face of competing demands, it is plausible that resources are more likely to
be directed at maintenance rather than at development. Hence, any attempts to trigger
development may encounter resistance. Ruminative triggers with its focus on symptoms
of the situation at hand may be closer to the catharsis and maintenance that one‘s battered
mind desires. As such, greater utility will be ascribed to ruminative triggers by trainees in
the rumination condition.
These observations remain conjectures at this point in time. However, there is
sufficient evidence to warrant that future research examine the effect of a developmental
climate on reflection and performance.
111
Reflective practices and habits. Another variable to include in future research is
actual reflective practices and habits acquired throughout the training course. This was
considered but not included in the current study as the focus was on demonstrating the
efficacy of micro-interventions introduced at the end of the training course for framing
one‘s reflections on training. One could argue that reflection can occur throughout the
entire duration of the training, not just at the end. Hence, including measures of reflective
habits and practices done throughout the entire course may have provided a more
accurate measure of change to better model the relationships between reflection and its
impact on leadership efficacy and Psycap development.
Developmental readiness and motivation to lead. Leadership development is
onerous, as is engaging in reflective practices. This is particularly so when the
environment is not conducive to such practices, or when one is not predisposed to
reflection. Hence, one must first be willing and ready for development in order to invest
energy and time into one‘s leadership development. In this study, trait reflection and
intermediate outcome measures such as quality of reflection were used to indicate the
motivation for reflection that occurred in the study. Motivation for reflection may also be
related to one‘s developmental readiness or one‘s motivation to lead. In this study, it was
not clear if trainees were motivated to develop themselves. Future research could include
this variable to help better understand the motivational dynamics behind reflection for
development.
Theoretical and Practical Contributions
This dissertation has provided empirical evidence for reflection with leadership
efficacy. However, the effect of reflection may not be uniformly beneficial across all
112
levels of leadership experience. The possible reasons for this interaction between
experience and reflection was discussed, of which one potential moderating variable
raised for future research was developmental climate.
Reflection is not enough. From a practitioner perspective, clearly, it is simplistic
to exhort reflection as a toolkit of leadership development without paying attention to the
level of leadership development in the leader, as well as attending to the climate of the
training milieu. In addition, other leader-related variables that needed to be considered
included the leader‘s personal readiness for development. Finally, attention needs to be
paid to encouraging personal habits of reflection in the leader. Habit creation implies that
the length of training required may necessarily be long, though the length of each
intervention need not be so.
Distinguishing skill development and identity development. Another
contribution this dissertation has made is to distinguish between leadership skill
development objectives versus facilitating psychological transitions in self-identities in
training. In this study, the trainees from the high leadership condition clearly emphasized
skill development rather than identity development. On the other hand, the training
provided to the trainees from the low leadership condition placed more emphasis on
identity development since these were civilians newly conscripted into the military.
Consequently, the data suggested that the higher quality of leader self-reflections were
experienced by those trainees from the low leadership experience condition. On the other
hand, although the trainees from the high leadership experience condition became more
skilled, they also reported poorer quality of leader self-reflections, presumably because
their identity development was not being addressed.
113
Motivation and identity. Leadership development practitioners need to attend to
not only ‗will‘ and ‗skill‘ factors, but also, as shown in this study, identity-transition
issues. For example, in an Academy of Management Learning & Education article,
Rousseau and McCarthy (2007) provided six principles to help managers become more
skillful in making better decisions based on evidence. Practitioners who adapt these
principles in managerial development programs will need to further attend to the
motivation of manager-trainees to want to apply these principles in real-life. While
extrinsic motivators can be provided as carrots for doing so, to trigger intrinsic
motivations, one would invariably need to see to the identity-development of these
managers.
Conclusion
This dissertation seeks to contribute towards closing the scientist-practitioner gap
in the area of reflection for leadership development. The oft-used leadership development
method of reflection was examined, firstly from two theoretical bases of educational
psychology and clinical psychology to derive two types of reflective triggers which were
demonstrated to work. Next, the reflective triggers were used in an applied training
setting to influence leadership development. In so doing, an unanticipated interaction
between level of leadership experience and the type of reflection was uncovered.
Nonetheless, empirical evidence was established to successfully link reflection with
psychological capital and leadership efficacy.
114
Figures
Figure 1: Hypothesized Model
Leadership Efficacy
Implicit Leadership
Leadership Self-Awareness
Control Variables:
- Trait Reflection
- Initial Leadership Efficacy
Leadership
Experience
Psycap
Type of Reflection
H1a
H1b
H1c
H1d
H2a H2b
H2c
Post Hoc
Post Hoc
H3a & b
H4a & b
H5a & b
H3-5 examined under low and
high leadership conditions
115
Figure 2: 3-stage model linking adaptive reflection to experience, development and performance
* Numbered Arrows indicate when the process of adaptive reflection occurs
** Boxes 1 & 2 contribute to development while Boxes 2 & 3 contribute to performance
2: ‗Transforming‘ Experience
Insight &
Knowledge
Solution / Course
of Action
Problem / Puzzle
of Practice
Experience &
Tacit Knowledge
1: ‗Grasping‘ Experience 3: Applying Experience
1 2
3
4
Heightened
Self-Awareness
116
Figure 3: Effect of Time 1 Predictors on Quality of Reflection at Time 2
PSYCAP at
Time 1
.08
.82*
.87*
.71*
.54* .28*
Leadership
Self-Awareness
at Time 2
Purpose
.52*
.80* .84*
Action - TA
Action - TF GLE at
Time 1
.94*
.88*
.55* .97*
.13*
Portrait
(Regulate)
Portrait
(Aware)
.62*
.70*
Q4
Q6 .81
.40
.44
.77
Q8
Q9 .55
.35
.67
.81
Q1
Q2
.74
.47 .78
.45
Paradigm
(Regulate)
Q17 Q21
.73 .58
Q19 Q20
.72 .47
Q18
.75
.47 .44 .49 .78 .66
Paradigm
(Aware)
Q12 Q15
.71 .53
Q14
.53
Q13
.77
.50 .41 .72 .72
Self/Thought
Q37 Q39
.71 .70
Q36 Q38
.77 .61
Q40
.74
.50 .45 .40 .63 .51 .52
Means
Q33 Q32
.69 .69
Q29 Q30
.66 .71
Q34
.70
.51 .57 .49 .52
Q24
Q26 .71
.47 .73
.72
Q23
Q25 .36
6
.36
.80
.54
Q28
Q27
.68
.64
.60
.54
Q22 .43
.80
.41
.50
.59
.58
.57
.54
Confidence
Q45
Q46
Q47
Q48
Q49
Q44 .68
.71
.77
.65
.64
.66
Hope
Q51 Q55
.65 .59
Q53 Q54
.77 .79
Q52
.52
.73 .41 .38 .58 .66
Resilience
Q57 Q60
.67 .66
Q59
.60
Q58
.52
.55 .73 .65 .57
Optimism
Q67
.51
.74
Q65
.81
.34
Q64
.81
.35
.12
.40
.61
.43
.47
.44
.41
.57
.52
.65
.44
.50
.52
Trait
Reflection
(Engage &
Need)
Q86
Q72
Q74
Q79
Q77
Q83 .69
.69
.65
.75
.59
.71
.63
Trait
Reflection
(Insight)
Q85
Q71
Q76
Q78
Q81
Q82 .61
.63
.78
.73
.76
.75
Q84
.23*
.18
.03
.77
117
Standardized beta weights as shown above. (Significant beta weights indicated by *)
Adequate Model Fit (CFI=.904, RMSEA=.036, SRMR=.056) Indicators of all latent factors are
not depicted in the figure.
118
Figure 4: Interaction of Military Leadership Experience with Type of Reflection on
Leadership Self-Awareness
119
Figure 5: Interaction of Military Leadership Experience with Type of Reflection on
Leadership Efficacy
120
Figure 6: Interaction of Leadership Experience with Type of Reflection on Quality of
Implicit Leadership Theories
121
Figure 7: Predictors of Generalized Leadership Efficacy at Time 2
Confi-
dence
Optimism
PSYCAP at
Time 2
.790*
.778*
.777*
.776*
.575
*
.169*
Leadership
Self-Awareness
at Time 2
Purpose
.50*
.85* .82*
Action - TA
Action - TF GLE at
Time 2
.928*
.918*
.692
*
.941*
.928*
.106*
Portrait
(Regulate)
Portrait
(Aware)
.65*
.63*
-.005
Q4
Q6 .81
.35
.44
.81
Q8
Q9 .58
.21
.65
.85
Q1
Q2
.82
.45 .80
.33
Paradigm
(Regulate)
Q17 Q21
.72 .62
Q19 Q20
.73 .51
Q18
.75
.49 .44 .47 .74 .61
Paradigm
(Aware)
Q12 Q15
.71 .56
Q14
.56
Q13
.75
.49 .44 .69 .69
Self/Thought
Q37 Q39
.73 .74
Q36 Q38
.77 .68
Q40
.77
.47 .40 .41 .54 .45 .55
Means
Q33 Q32
.67 .76
Q29 Q30
.71 .69
Q34
.76
.42 .50 .51 .43
Q24
Q26 .63
.36 .80
.76
Q23
Q25 .35
.43
.81
.61
Q28
Q27
.79
.67
.55
.38
Q22 .43
.75
Q45
Q46 .36
.38
Q47
Q48 .58
.50
Q49 .45
Q44 .45 .74
.79
.80
.71
.65
.75
Hope
Q51 Q55
.64 .67
Q53 Q54
.81 .84
Q52
.57
.67 .35 .29 .59 .55
Resilience
Q57 Q60
.75 .70
Q59
.80
Q58
.61
.44 .63 .37 .51
Q67
.60
.65
Q65
.88
.23
Q64
.86
.26
122
Standardized beta weights as shown above. (Significant beta weights indicated by *)
Adequate Model Fit (CFI=.900, RMSEA=.046, SRMR=.055) Indicators of all latent factors are
not depicted in the figure.
123
Tables
Table 1: Events that provide highest proportions of developmental triggers
Event % of Developmental Triggers
Route March 35.01
Field Camp 29.22
Navigation Exercise 12.66
1st day in course 5.18
Others 17.93
Total 100.00
124
Table 2: Coding Frame for evaluating reflections
Self-Awareness Self-Regulation and
Self-monitoring
Leadership Purpose and Values 9% 15%
Leader self-image, self-portrait 14% 6%
Leadership assumptions, operating paradigms 16% 3%
Others (not captured in the coding frame) 37%
125
Table 3: EFA results
Model Chi Sq df sig RMSEA RMSEA
< .06?
Chi-Sq
Diff
df
Diff
Chi-Sq Diff
Sig
1-factor 909.28 189 <.001 0.089 No - - -
2-factors 632.84 169 <.001 0.076 No 276.44 20 <.0001
3-factors 430.25 150 <.001 0.063 No 202.59 19 <.0001
4-factors 298.73 132 <.001 0.051 Yes 131.52 18 <.0001
5-factors 220.78 115 <.001 0.044 Yes 77.95 17 <.0001
6-factors* - - - - - - - -
*Did not converge
126
Table 4: EFA item loadings for Quality of Reflection Scale in Pilot Sample
Item F1 F2 F3 F4 F5
T1Q1 0.508
T1Q2 0.592
T1Q3 0.498
T1Q4 0.448
T1Q5 0.98
T1Q7 0.333
T1Q8 0.549
T1Q9 0.732
T1Q10 0.342 0.321
T1Q11 0.613
T1Q12 0.717
T1Q13 0.700
T1Q14 0.416
T1Q15 0.524
T1Q16 0.447 0.431
T1Q17 0.623
T1Q18 0.618
T1Q19 0.733
T1Q20 0.773
T1Q21 0.512
127
*Item 6 failed to load on any latent factor. Item-Latent Factor correlation lower than .3 were
filtered.
128
Table 5: EFA Item Loadings for Trait Reflectiveness Scale in Pilot Sample
ITEM F1 F2 F3 F4
T1Q71r I‘m often confused about the way that I really feel
about things
0.682
T1Q72 It is important for me to evaluate the things that I do 0.748
T1Q74 I am very interested in examining what I think about 0.760
T1Q76 I‘m often aware that I‘m having a feeling, but I often
don‘t quite know what
0.652 -0.348
T1Q77 I frequently examine my feelings 0.392 -0.314 0.816
T1Q78r My behavior often puzzles me 0.730
T1Q79 It is important to me to try to understand what my
feelings mean
0.656 -0.339
T1Q81 Thinking about my thoughts make me more confused 0.729
T1Q82 I have a definite need to understand the way that my
mind works
0.328 -0.418 0.246
T1Q83 I frequently take time to reflect on my thoughts 0.697
T1Q84r Often I find it difficult to make sense of the way I
feel about things
0.738
T1Q85 It is important to me to be able to understand how my
thoughts arise
0.422 -0.759
T1Q86 I often think about the way I feel about things 0.396 -0.477 0.324
Item-Latent Factor correlation lower than .3 were filtered.
129
Table 6: Descriptive Statistics for Manipulation Check Item 1
Low Leadership Experience High Leadership Experience
N Mean Sd N Mean Sd
Reflection 148 3.47 .852 167 3.47 .883
Rumination 164 3.25 .868 90 3.39 .817
Control 148 3.63 .794 74 3.80 1.020
130
Table 7: Descriptive Statistics for Manipulation Check Item 2
Low Leadership Experience High Leadership Experience
N Mean Sd N Mean Sd
Reflection 148 4.23 .948 167 4.64 .793
Rumination 163 4.12 .996 90 4.22 1.003
Control 148 4.48 .778 74 4.46 .996
131
Table 8: Descriptive Statistics for Manipulation Check Item 3
Low Leadership Experience High Leadership Experience
N Mean Sd N Mean Sd
Reflection 147 2.99 1.047 167 3.27 1.138
Rumination 160 3.36 1.095 90 3.04 1.038
Control 147 2.76 .894 74 2.49 1.337
132
Table 9: Descriptive Statistics for Variables in Basic Military Course Study Sample (Low Leadership Experience)
Variable Reflection Group Rumination Group Control Group Overall
N Mean S.D. N Mean S.D. N Mean S.D. N Mean S.D.
T2Reflection 180 3.583 .495 167 3.516 .4779 150 3.579 .474 497 3.560 .483
T2GLE 180 6.295 1.323 167 6.183 1.190 150 6.432 1.327 497 6.299 1.282
T2Psycap 180 4.569 .631 167 4.462 .669 150 4.568 .639 497 4.533 .647
T2TraitR 180 4.318 .763 167 4.466 .687 150 4.595 .676 497 4.451 .720
T2NGSE 180 4.581 .766 167 4.590 .792 150 4.710 .747 497 4.623 .770
T2GOApprch 180 4.631 .740 167 4.754 .958 150 4.796 .757 497 4.722 .826
T2GOAvoid 180 4.214 .724 167 4.394 .749 150 4.398 .727 497 4.330 .737
T2GOLearn 180 4.998 .948 167 5.087 1.019 150 5.193 .861 497 5.087 .949
ILT 180 3.514 1.470 167 2.872 1.065 150 3.205 1.358 497 3.183 1.324
133
Table 10: Descriptive Statistics for Variables in Leadership Course Study Sample (High Leadership Experience)
Variable Reflection Group Rumination Group Control Group Overall
N Mean S.D. N Mean S.D. N Mean S.D. N Mean S.D.
T1Reflection 162 3.566 0.411 79 3.662 0.413 71 3.607 0.385 312 3.600 0.406
T2Reflection 162 3.552 0.435 86 3.711 0.455 71 3.616 0.398 319 3.609 0.437
T1GLE 161 6.453 0.994 80 6.792 1.019 70 6.554 0.988 311 6.563 1.006
T2GLE 162 6.469 1.032 86 6.890 0.968 71 6.669 0.969 319 6.627 1.014
T1Psycap 161 4.604 0.537 81 4.651 0.594 70 4.633 0.535 312 4.623 0.550
T2Psycap 162 4.578 0.569 85 4.730 0.608 71 4.666 0.522 318 4.638 0.572
T1TraitR 161 4.528 0.627 81 4.755 0.620 70 4.630 0.633 312 4.610 0.632
T2TraitR 162 4.578 0.639 85 4.679 0.646 71 4.636 0.634 318 4.618 0.639
T1NGSE 160 4.800 0.666 81 4.858 0.712 70 4.767 0.678 311 4.808 0.679
T2NGSE 162 4.794 0.642 85 4.914 0.719 71 4.873 0.676 318 4.844 0.671
T1GOApprch 160 5.316 0.897 81 5.506 0.859 70 5.389 0.927 311 5.382 0.895
T2GOApprch 162 5.276 0.860 85 5.435 0.894 71 5.454 0.860 318 5.358 0.871
134
T1GOAvoid 160 4.360 0.973 81 4.634 1.197 70 4.471 0.976 311 4.457 1.039
T2GOAvoid 162 4.389 1.033 85 4.478 1.196 71 4.446 0.891 318 4.426 1.048
T1GOLearn 160 5.298 0.807 81 5.483 0.700 70 5.329 0.733 311 5.353 0.765
T2GOLearn 162 5.210 0.755 85 5.476 0.780 71 5.243 0.814 318 5.288 0.781
T2ILT 162 4.804 1.789 85 4.325 1.653 71 4.541 1.586 318 4.615 1.716
135
Table 11: Skewness and Kurtosis Statistics for Variables in Basic Military Course Study Sample (Low Leadership Experience)
Variable Reflection Group Rumination Group Control Group Overall
Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis
T2Reflection .330 -.048 -.120 .701 -.609 1.222 -.079 .523
T2GLE -.312 -.184 -.619 .214 -.811 1.413 -.535 .403
T2Psycap -.429 .587 -.530 .013 -.375 .463 -.457 .336
T2TraitR .375 -.806 -.276 -.185 .007 -.672 .035 -.671
T2NGSE -.285 -.128 -.811 1.121 -.801 1.752 -.619 .775
T2GOApprch .314 1.252 -.572 1.021 .118 .485 -.165 1.072
T2GOAvoid .582 .600 .119 -.120 .405 .366 .360 .167
T2GOLearn -.360 -.048 -.719 .943 -.287 .417 -.505 .757
136
Table 12: Skewness and Kurtosis Statistics for Variables in Leadership Course Study Sample (High Leadership Experience)
Variable Reflection Group Rumination Group Control Group Overall
Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis
T1Reflection -0.099 0.581 0.367 -0.162 0.500 0.454 0.135 0.404
T2Reflection 0.125 -0.294 0.581 0.091 0.032 -0.444 0.258 -0.050
T1GLE -0.361 -0.066 -0.365 0.362 -0.812 0.491 -0.431 0.120
T2GLE -0.168 -0.050 -0.279 -0.542 -0.397 -0.530 -0.256 -0.277
T1Psycap -0.337 0.426 -0.363 0.643 -0.344 0.600 -0.336 0.497
T2Psycap -0.068 0.048 0.030 -0.546 -0.456 0.779 -0.087 -0.033
T1TraitR 0.232 -0.375 -0.015 -0.532 -0.235 -0.114 0.058 -0.436
T2TraitR 0.125 -0.402 0.150 -0.669 0.103 -0.643 0.126 -0.539
T1NGSE -0.599 1.647 -0.758 1.086 -0.422 0.616 -0.593 1.151
T2NGSE -0.008 -0.430 -0.318 -0.556 -0.143 -0.300 -0.115 -0.488
T1GOApprch -0.099 -0.440 -0.264 -0.004 0.076 -0.234 -0.101 -0.330
T2GOApprch 0.004 0.186 -0.119 -0.312 -0.345 -0.238 -0.099 -0.125
137
T1GOAvoid 0.292 0.231 -0.032 -0.307 -0.116 -0.166 0.149 -0.042
T2GOAvoid -0.255 1.119 -0.153 0.306 -0.055 0.041 -0.180 0.754
T1GOLearn -0.045 -0.112 0.459 -0.329 -0.325 0.841 -0.032 0.107
T2GOLearn 0.157 0.005 0.240 -0.647 0.337 -0.010 0.225 -0.179
138
Table 13: Summary of Hypotheses
Hypothesis Analysis Results
H1a: The level of Psycap is positively
correlated with prior leadership experience.
H1b: The level of leadership efficacy is
positively correlated with prior leadership
experience.
H1c: The level of leadership self-awareness is
positively correlated with prior leadership
experience.
H1d: The quality of implicit leadership theories
is positively correlated with prior leadership
experience.
A ONE-WAY ANOVA of level of prior
leadership experience on outcomes of
Psycap, leadership efficacy, leadership
self-awareness and implicit leadership
theories
1a & 1c not supported
1b & 1d supported. Higher levels of
leadership experience are associated
with higher levels of leadership efficacy
and quality of implicit leadership
theories.
H2a: The relationship between leadership
efficacy and leadership experience is moderated
by type of reflection triggers.
H2b: The relationship between leadership self-
awareness and leadership experience is
moderated by type of reflection triggers.
H2c: The relationship between quality of
implicit leadership theories and leadership
experience is moderated by type of reflection
triggers.
A 3 (reflection, rumination, control) by
2 (low vs high leadership experience)
MANCOVA on GLE, SA & implicit
leadership (using SPSS GLM) with trait
self-efficacy and trait reflection as
controls. For this, to explore only the
simple effects of leadership experience
on the outcomes with the controls
present.
H2a & b supported.
H2c not supported.
Interaction of leadership experience and
type of reflection trigger observed for
leadership efficacy and leadership self-
awareness but not for quality of implicit
leadership theories.
139
H3a: Recalling and reflecting positively on any
personal leadership experience increases
subsequent leadership efficacy.
H3b: Recalling and reflecting negatively on any
personal leadership experience decreases
subsequent leadership efficacy.
A ONE-WAY ANOVA conducted
separately on low and high leadership
samples for type of reflection triggers on
leadership efficacy, leadership self-
awareness and implicit leadership
theories with post-hoc analyses
Low leadership experience: Not
supported. Expected direction but not
significant.
High leadership experience: Not
supported. Significant difference but
rumination condition scores higher.
H4a: Recalling and reflecting on past
experience results in greater self-awareness.
H4b: Ruminative recall of past experience
results in greater self-awareness compared with
adaptive recall.
As above Low leadership experience: Not
supported. Expected direction but not
significant.
High leadership experience: Not
supported. Significant difference but
rumination condition scores higher.
H5a: Adaptive reflection results in positive
aspects of one‘s implicit leadership theories
becoming more easily explicated.
H5b: Maladaptive reflection results in negative
aspects of one‘s implicit leadership theories
becoming more easily explicated.
As above Supported under both low and high
leadership conditions.
140
Appendices
Appendix A: Sequence of Activities at Time 1 & 2
Activity Reflection Gp Rumination Gp Control Gp
Intervention 1. 7-item Reflection Exercise: Field Camp**
1. 7-item Rumination
Ex
1. Sudoko puzzle 1 (3x3)
2. Sudoko puzzle 2 (9x9)
Survey 1. 3-item Manipulation check items**
2. 5-item Developmental Goal Setting Exercise*
3. 5-item Implicit Leadership Theory Sentence Completion Ex**
4. 21-item Quality of Reflection Outcome
5. 22-item Generalized Leadership Efficacy
6. 24-item Psychological Capital
7. 20-item Trait Reflectiveness scale*
8. 8-item Generalized Self Efficacy scale*
9. Rated Quality of Goal setting**
10. 4-item Demographics
same same
* Only for Time 1 ** Only for Time 2
141
Appendix B: Measures
Measure 1: Quality of Reflection
The scale anchors are 1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Frequently, 5 = All the
time.
q1 I can cite ways which my personal values reinforces the SAF core values
q2 I can cite ways that my personal values and the SAF core values are different but still
reinforcing
q3 I can cite challenges where I was forced to choose between my personal values and
the SAF core values*
q4 I know how others view my strengths as a leader
q5 I know how others view my weaknesses as a leader*
q6 I know where people differ in opinions regarding my leadership actions
q7 I ask for feedback on my leadership impact on others*,
q8 I am conscious of my behavior when leading others
q9 I control my behavior as a leader in the presence of others
q10 I consider how my emotions can affect others*
q11 I know when to consult with others regarding how best to address challenges*
q12 I understand the reasons behind how i behave as a leader
q13 I know why I feel the way I do as a leader
q14 I know when I am getting too involved as a leader with directing others
q15 I know how my feedback impacts the motivation of others
q16 I know how to adjust my leadership style to energise others to action*
q17 I examine whether my reasons for how I behave as a leader are still valid
142
q18 I consider why I feel the way I do as a leader before I act
q19 I revise the way I look at myself as a leader
q20 I challenge my own firmly held ideas about leadership
q21 I look to adapt how I do things a leader
* Items removed from further usage after EFA and CFAs
143
Measure 2: Generalized Leadership Efficacy
Directions: For each item below, indicate your level of confidence in your ability to
accomplish each task or activity as a leader in your organization in the future. Use the
following scale to indicate your level of confidence and mark your answer to the right
next to each item. A score of 10 represents 100% confidence, whereas a score of 0 means
no confidence at all.
0--------1-------2------3------4------5------6------7------8------9------10
Not at all Moderately Totally
Confident Confident Confident
Actions
q22 develop agreements with followers to enhance their participation.
q23 inspire followers to go beyond their self-interests for the greater good.
q24 get my followers to meet the requirements we have set for their work.
q25 inspire followers to perform beyond their expectations.
q26 come up with the rewards and punishments that will work best with my
followers.
q27 get followers to re-examine their basic beliefs and assumptions.
q28 coach followers to assume greater responsibilities for leadership.
Means
q29 count on my leaders to support high standards of ethical conduct.
q30 go to my superiors for advice to develop my leadership.
q31 effectively lead working within the boundaries of my organization's policies.*
q32 rely on my leaders to come up with ways to stimulate my creativity.
q33 count on others to give me the guidance I need to complete work assignments
q34 rely on my organization to provide the resources needed to be effective.
q35 rely on my peers to help solve problems.*
Self Thought
q36 adapt my thinking to a broad range of unique leadership challenges.
q37 motivate myself to set goals that are achievable.
q38 remain steadfast to my core beliefs when I'm challenged.
q39 motivate myself to perform at levels that inspire others to excellence.
q40 develop detailed plans to accomplish complex missions.
q41 accomplish the targeted goals set by my superiors.*
q42 determine the objectives needed to complete our project goals.*
q43 distinguish the ethical components of problems/dilemmas.*
144
GLE Scoring Scheme for Subscales:
Self/Thought—Transactional: Items 37, 40, 41, 42
Self/Thought—Transformational: Items 36, 38, 39, 43
Action—Transactional: Items 22, 24, 26
Action—Transformational: Items 23, 25, 27, 28
Means—Transactional: Items 31, 33, 34
Means—Transformational: Items 29, 30, 32, 35
* Items removed from further usage after CFAs
145
Measure 3: Psycap Scale (4 Components)
1. Job Role Based Self-Efficacy (adapted from Parker, 1998)
1 = not at all confident, 2 = not confident, 3 = somewhat not confident, 4 = somewhat
confident, 5 = confident, 6 = very confident
How confident would you feel....
q44 Analyzing a long-term problem to find a solution?
q45 Representing your work area in meetings with senior management?
q46 Contributing to discussions about the company's strategy?
q47 Helping to set targets/goals in your work area?
q48 Contacting people outside the company (e.g., suppliers, customers) to discuss
problems?
q49 Presenting information to a group of colleagues?
2. Job Role Based Hope (Adapted from Snyder, 1996)
1 = Strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = somewhat agree, 5 =
agree, 6 = strongly agree
q50 If I should find myself in a jam at work, I could think of many ways to get out of
it.*
q51 At the present time, I am energetically pursuing my work goals.
q52 There are lots of ways around any problem that I am facing now at work.
q53 Right now I see myself as being pretty successful at work.
146
q54 I can think of many ways to reach my current work goals.
q55 At this time, I am meeting the work goals that I have set for myself.
3. Job Role Based Resiliency (Adapted from Wagnild, G., & Young, H., 1993).
1 = Strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = somewhat agree, 5 =
agree, 6 = strongly agree
q56 When I have a setback at work, I have trouble recovering from it and moving on.*
q57 I usually manage difficulties one way or another at work.
q58 I can be ―on my own‖ so to speak at work if I have to.
q59 I usually take stressful things at work in stride.
q60 I can get through difficult times at work because I've experienced difficulty before.
q61 I feel I can handle many things at a time at this job.*
4. Job Role Based Optimism (Adapted from Scheier and Carver, 1994)
1 = Strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = somewhat agree, 5 =
agree, 6 = strongly agree
q62 When things are uncertain for me at work I usually expect the best.*
q63 If something can go wrong for me work-wise it will.*
q64 I always look on the bright side of things regarding my job.
q65 I‘m optimistic about what will happen to me in the future as it pertains to work.
q66 In this job, things never work out the way I want them to.*
q67 I approach this job as if ―every cloud has a silver lining‖.
* Items removed from further usage after CFAs
147
Measure 4: Trait Reflectiveness Scale
The scale used is the Self-reflection and Insight Scale (Grant et al., 2002). This
20-item scale has 12 items measuring self-reflection ("the inspection and evaluation of
one's thoughts, feelings and behavior") and 8 items measuring insight ("the clarity of
understanding of one's thoughts, feelings, and behavior"; Grant et al., 2002, p. 821).
q68 I don't often think about my thoughts* (R) (E)
q69 I am not really interested in analyzing my behavior* (R) (N)
q70 I am usually aware of my thoughts* (I)
q71 I'm often confused about the way that I really feel about things (R) (I)
q72 It is important for me to evaluate the things that I do (N)
q73 I usually have a very clear idea about why I've behaved in a certain way* (I)
q74 I am very interested in examining what I think about (N)
q75 I rarely spend time in self-reflection* (R) (E)
q76 I'm often aware that I'm having a feeling, but I often don't quite know what (R) (I)
q77 I frequently examine my feelings (E)
q78 My behaviour often puzzles me (R) (I)
q79 It is important to me to try to understand what my feelings mean (N)
q80 I don't really think about why I behave in the way that I do* (R) (E)
q81 Thinking about my thoughts makes me more confused (R) (I)
q82 I have a definite need to understand the way that my mind works (N)
q83 I frequently take time to reflect on my thoughts (E)
q84 Often I find it difficult to make sense of the way I feel about things (R) (I)
148
q85 It is important to me to be able to understand how my thoughts arise (N)
q86 I often think about the way I feel about things (E)
q87 I usually know why I feel the way I do (I)
E = Engagement in self-reflection: N = Need for self-reflection: I = Insight: R = Reverse
scored
* These items were excluded from subsequent use after the EFA and CFAs
Scoring Instructions
The scoring process is very simple. Summed scores are used. There is no scaling or
scale transformation required other than basic reverse scoring for four items.
Step 1.
Reverse score those items marked (R).
An original score of "1" would become"6"; "2" would become "5"; "3" becomes "4" and
visa versa.
Step 2.
Sum the scores for each subscale
149
Measure 5: New General Self-Efficacy Scale (Chen, Gully & Eden, 2001)
Anchors are 1 = Strongly Disagree, 2 = Disagree, 3 = Somewhat Disagree, 4 = Somewhat
Agree, 5 = Agree, 6 = Strongly Agree
q88 I will be able to achieve most of the goals that I have set for myself*
q89 When facing difficult tasks, I am certain that I will accomplish them
q90 In general, I think that I can obtain outcomes that are important to me
q91 I believe I can succeed at most any endeavor to which I set my mind
q92 I will be able to successfully overcome many challenges
q93 I am confident that I can perform effectively on many different tasks
q94 Compare to other people, I can do most tasks very well*
q95 Even when things are tough, I can perform quite well
* These items were removed from further usage after the CFAs.
150
Manipulation Check items
1. How did you feel about the event/activity you were told to think about?
1 – very negative
2 – negative
3 – neutral
4 – positive
5 – very positive
2. Your thinking about the event/activity made you feel
1 – very depressed
2 – quite depressed
3 – depressed
4 – slightly depressed
5 – not depressed at all
1. Your thinking about the event/activity made you feel
1 – very hopeful
2 – quite hopeful
3 – hopeful
4 – slightly hopeful
5 – not hopeful at all
151
Outcome Measure 1: Sentence Completion Exercise (Implicit Leadership
Theory)
Please complete the phrase below with your own inputs. Please be sure to write clearly.
Explain your inputs as if the person who is going to read this does not know anything
about leadership at all.
1. To me, leadership is about
2. To be a leader, one must be
3. As a leader, one needs to take into consideration
4. My leadership is characterized by
5. The most effective leader is one who is
152
Outcome Measure 2: 5-Item Goal Setting Exercise
1. If I were to allocate how I spend my free time in future, I would allocate ___% to
developing myself as a leader.
2. My plan to develop myself as a leader is as follows:
3. The reasons why I developed my plan in the above manner is because
4. How much effort will you put to developing yourself as a leader? ___%
5. The main obstacles I foresee to me improving as a leader are
153
Appendix C: Prompts to prime Reflection And Rumination
Reflection Prompts Rumination Prompts
What were some of the problems you
encountered at the field camp?
What did you do to resolve the
problem?
What values (if any) were at play that
caused the problem?
What were some of the problems you
encountered at the field camp?
Describe how things went wrong.
What values (if any) were at play that
caused the problem?
What was the most meaningful part of
the field camp for you?
What was it about the event that made
it meaningful for you?
What part of the field camp was out of
your comfort zone? i.e. either made you
uncomfortable or anxious?
What limitations did it reveal about
yourself as a person and as a leader?
What did you learn about yourself
during the field camp?
How have your insights make you a
better person and a leader?
Do you think in future you will be
exposed in a similar way again?
Where do you think you need to
improve as a person and as a leader?
154
References
Allport, G. W. (1954). The nature of prejudice. Cambridge, Mass.,: Addison-Wesley Pub.
Co.
Argote, L. (2005). Organizational learning : creating, retaining and transferring
knowledge. New York, ny: Springer.
Argyris, C. (1999). On organizational learning (2 ed.). UK: Blackwell.
Argyris, C., & Schön, D. (1978). Organizational learning: A theory of action perspective.
Reading, Mass: Addison Wesley.
Ashford, S. J., & Tsui, A. S. (1991). Self-regulation for managerial effectiveness: The
role of active feedback seeking. Academy of Management Journal, 34, 251-280.
Avolio, B. J., Gardner, W. L., Walumbwa, F. O., Luthans, F., & May, D. R. (2004).
Unlocking the mask: A look at the process by which authentic leaders impact
follower attitudes and behaviors. The Leadership Quarterly, 15, 801-823.
Avolio, B. J., & Luthans, F. (2006). The high impact leader : moments matter in
accelerating authentic leadership development. New York: McGraw Hill.
Baltes, P. B., & Smith, J. (1990). Towards a psychology of wisdom and its ontogenesis.
In R. J. Sternberg (Ed.), Wisdom: Its nature, origins and development (pp. 87-
120). Cambridge, UK: Cambridge University Press.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Bandura, A. (2000). Cultivate self-efficacy for personal and organizational effectiveness.
In E. A. Locke (Ed.), The Blackwell handbook of principles of organizational
behavior (pp. 154-196). New York: Guilford.
Bandura, A. (2001). Guide for Constructing Self-Efficacy scales (Revised March 2001).
Available from Frank Pajares, Emory University.
Bettin, P. J., & Kennedy, J. J. K. (1990). Leadership experience and leader performance:
Some empirical support at last. The Leadership Quarterly, 1(4), 219.
Boal, K. B., & Bryson, J. M. (1988). Charismatic leadership: A phenomenological and
structural approach. In J. G. Hunt, B. R. Baliga, H. P. Dachler & C. A.
Schriesheim (Eds.), Emerging leadership vistas (pp. 11-28). Lexington, MA:
Heath.
Bolton, G. (1999). Reflection through the looking-glass. The story of a course of writing
as reflexive practitioner. Teaching in Higher Education, 4(2), 193-212.
Brown, B. K., & Campion, M. A. (1994). Biodata phenomenology: Recruiters'
perceptions and use of biographical information in resume screening. Journal of
Applied Psychology, 79, 897-908.
Brown, K. W., & Ryan, R. M. (2003). The benefits of being present: Mindfulness and its
role in psychological well-being. Journal of Personality & Social Psychology,
84(4), 822-848.
Bruner, J. (1996). Frames of thinking: Ways of making meaning. In D. Olson & N.
Torrance (Eds.), Modes of Thought (pp. 93-105). New York: Cambridge
University Press.
Cameron, G., Hayes, V., E., & Wren, A., Mah. (2000). Using reflective process in
community-based participatory action research. Reflective Practice, 1(2), 215-
230.
155
Cannella, J. A. A., & Rowe, W. G. (1995). Leader capabilities, succession, and
competitive context: A study of professional baseball teams. The Leadership
Quarterly, 6(1), 69.
Cantor, N., & Mischel, W. (1977). Traits as prototypes: Effects on recognition memory.
Journal of Personality and Social Psychology, 35, 38-48.
Carver, C., S., & Scheier, M., F. (1982). Control theory: A useful conceptual framework
for personality-social, clinical and health psychology. Psychological Review, 97,
19-35.
Carver, C. S., & Scheier, M. F. (1990). Origins and functions of positive and negative
affect: A control-process view. Psychological Review, 97, 19-35.
Catell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral
Research, 1, 245-276.
Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a New General Self-Efficacy
Scale. Organizational Research Methods, 4(1), 62-83.
Clyde, L., A., & Klobas, J., E. (2000). Shared reflection by email: Its role in new
information technology education and training. Education for Information, 18,
273-287.
Conger, J. A. (1989). Leadership: The Art of Empowering Others. Academy of
Management Executive, 3(1), 17.
Copeman, G. H. (1971). The chief executive and business growth; a comparative study in
the United States, Britain and Germany. London, New York,: Leviathan House.
Cramer, K. M. (2000). Comparing the relative fit of various factor models of the self-
consciousness scale in two independent samples. Journal of Personality
Assessment, 75, 295-307.
Creed, A. T., & Funder, D. C. (1998). The two faces of private self-consciousness: Self
report, peer-report and behavioral correlates. European Journal of Personality,
12(411-431).
Day, D. B. (2000). Leadership development: A review in content. The Leadership
Quarterly, 11, 581-614.
De Pree, M. (2004). Leadership is an art. New York: Currency.
Denny, N., W., Dew, J., R., & Kroupa, S., L. (1995). Perceptions of wisdom: What is it
and who has it? Journal of Adult Development, 2, 37-47.
Dewey, J. (1933). How we think: A restatement of the relation of reflective thinking to the
educative process. . New York: D. C. Heath and Company.
DiTomaso, N., & Hooijberg, R. (1996). Diversity and the demands of leadership.
[STUDY TYPE: Theoretical. No propositions. Kinda of a setup piece.]. The
Leadership Quarterly, 7(2), 163-187.
Duval, S., & Wicklund, R. A. (1972). A theory of objective self awareness. New York,:
Academic Press.
Dweck, C. S., Chiu, C., & Hong, Y. (1995). Implicit theories and their role in judgments
and reactions: A world from two perspectives. Psychological Inquiry, 6, 267-285.
Dweck, C. S., Hong, Y., & Chiu, C. (1993). Implicit theories: Individual differences in
the likelihood and meaning of dispositional inference. Personality & Social
Psychology Bulletin, 19, 644-656.
156
Eden, D., & Leviatan, U. (1975). Implicit leadership theories as a determinant of the
factor structure underlying supervisory behavior scales. Journal of Applied
Psychology, 60, 736-741.
Edmondson, A. (1999). Psychological safety and learning behavior in work teams.
Administrative Science Quarterly, 44, 350-385.
Edmondson, A. (2002). The local and variegated nature of learning in organizations: A
group-level perspective. Organizational Science, 13, 128-146.
Ellis, S., & Davidi, I. (2005). After-Event Reviews: Drawing Lessons From Successful
and Failed Experience. Journal of Applied Psychology, 90(5), 857-871.
Engle, E. M., & Lord, R. G. (1997). Implicit theories, self-schemas and leader-member
exchange. Academy of Management Journal, 40, 988-1010.
Erdfelder, E., Faul, F., & Buchner, A. (1996). GPOWER: A general power analysis
program. Behavior Research Methods, Instruments, & Computers, 28, 1-11.
Fairhurst, G. T., & Sarr, R. A. (1996). The art of framing: Managing the language of
leadership. San Francisco, CA: Jossey-Bass.
Farbrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating
the use of exploratory factor analysis in psychological research. Psychological
Methods, 4, 272-299.
Fejfar, M. C., & Hoyle, R. H. (2000). Effect of Private Self-Awareness on Negative
Affect and Self-Referent Attribution: A Quantitative Review. Personality &
Social Psychology Review, 4(2), 132.
Fiedler, F. E. (1970). Leadership Experience and Leader Performance - Another
Hypothesis Shot to Hell. Organizational Behavior & Human Performance, 5(1),
1.
Fiedler, F. E. (1972). Predicting the effects of leadership training and experience from the
contingency model. Journal of Applied Psychology, 56(2), 114.
Fiedler, F. E. (1995). The effect and meaning of leadership experience: A review of
research and preliminary model. In K. E. Clark, M. B. Clark & D.P.Campbell
(Eds.), Impact of leadership. Greensboro, NC: Center for Creative Leadership.
Fiedler, F. E. (2002). The curious role of cognitive resources in leadership. In R. Riggio,
E., S. E. Murphy & F. J. Pirozzolo (Eds.), Multiple intelligences and leadership
(pp. 91-104). Mahwah, New Jersey: Lawrence Erlbaum Associates.
Fiedler, F. E., & Garcia, J. E. (1987). New approaches to effective leadership : cognitive
resources and organizational performance. New York: Wiley.
Fiedler, F. E., McGuire, M. A., & Richardson, M. (1989). The role of intelligence and
experience in successful group performance. Journal of Applied Sport
Psychology, 1(2), 132-149.
Fiske, S. T., & Taylor, S. E. (1984). Social Cognition. Reading, MA: Addison-Wesley.
Foti, R. J., & Luch, C. H. (1992). The influence of individual differences on the
perception and categorization of leaders. Special Issue: Individual differences and
leadership: II. [STUDY TYPE: Theoretical]. The Leadership Quarterly, 3(1), 55-
66.
Freedman, L. S. (1982). Tables of the number of patients required in clinical trials using
the logrank test. Statistics in Medicine, 1, 121-129.
Gabarro, J. J. (1987). The dynamics of taking charge. Boston, Mass.: Harvard Business
School Press.
157
Gardner, W. L., Avolio, B. J., Luthans, F., May, D. R., & Walumbwa, F., O. (2005).
"Can you see the real me?" A self-based model of authentic leader and follower
development. The Leadership Quarterly, 16(3), 343-372.
Gibson, C., & Vermeulen, F. (2003). A healthy divide: Subgroups as a stimulus for team
learning behavior. Administrative Science Quarterly, 48, 202-239.
Glorfeld, L. W. (1995). An improvement on Horn's parallel analysis methodology for
selecting the correct number of factors to retain. Educational and Psychological
Measurement, 55, 377-393.
Grant, A., M., Franklin, J., & Langford, P. (2002). The self-reflection and insight scale: A
new measure of private self-consciousness. Social Behavior & Personality, 30(8),
812-836.
Greene, M. (1988). Reflection and passion in teaching. Journal of Curriculum and
Supervision, 2(1), 68-81.
Hannah, S., Avolio, B. J., Chan, A., & Mhatre, K. (2005). Agentic Leadership Efficacy
Scale: Validation of scale and prediction of performance. Gallup Leadership
Institute, University of Nebraska-Lincoln.
Hannah, S., Woolfolk, R. L., & Lord, R. G. (2009). Leader self-structure: A framework
for positive leadership. Journal of Organizational Behavior, 30, 269-290.
Harter, N. (2002). Authenticity. In C. R. Snyder & S. Lopez (Eds.), Handbook of positive
psychology. Oxford: Oxford University Press.
Hedlund, J., Forsythe, G. B., Horvath, J. A., Williams, W. M., Snook, S., & Sternberg, R.
J. (2003). Identifying and assessing tacit knowledge: understanding the practical
intelligence of military leaders. The Leadership Quarterly, 14(2), 117.
Heifetz, R. A. (1994). Leadership without easy answers. Cambridge, Mass.: Belknap
Press of Harvard University Press.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological
Review, 94, 319-340.
Hirst, G., Mann, L., Bain, P., Pirola-Merlo, A., & Richver, A. (2004). Learning to lead:
the development and testing of a model of leadership learning. The Leadership
Quarterly, 15(3), 311-327
Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis.
Psychometrika, 32, 179-185.
House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (2004). Culture,
leadership, and organizations: The GLOBE study of 62 cultures. Thousand Oaks,
CA: Sage.
Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity
to underparameterized model misspecification. Psychological Methods, 3, 424-
453.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indices in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling, 6, 1-55.
Hunt, J. G., Boal, K. B., & Sorenson, R. L. (1990). Top management leadership: Inside
the black box. [STUDY TYPE: Theoretical]. The Leadership Quarterly, 1(1), 41-
65.
Ingram, R. E. (1990). Self-focused attention in clinical disorders: Review and a
conceptual model. Psychological Bulletin, 107, 156-176.
158
Jung, D. I., & Avolio, B. (2000). Opening the black box: an experimental investigation of
the mediating effects of trust and value congruence on transformational and
transactional leadership. Journal of Organizational Behavior, 21(8), 949-964.
Kaiser, H. F. (1960). The application of electronic computers to factor analysis.
Educational and Psychological Measurement, 20, 141-151.
Kegan, R. (1994). In over our heads: The mental demands of modern life. Cambridge,
Mass: Harvard University Press.
Kegan, R. (2000). What "form" transforms? A constructive-developmental approach to
transformative learning. In J. Mezirow & Associates (Eds.), Learning as
transformation (pp. 35-69). San Francisco: Jossey-Bass.
Keller, T. (1999). Images of the familiar: Individual differences and implicit leadership
theories. [STUDY TYPE: Empirical (Survey)]. The Leadership Quarterly, 10(4),
589-607.
Keller, T. (2003). Parental images as a guide to leadership sensemaking: an attachment
perspective on implicit leadership theories. The Leadership Quarterly, 14(2), 141.
Kelley, R. E. (1992). The power of followership: Bantam Dell.
Kelley, R. E. (1998). Followership in a leadership world. In L. C. Spears (Ed.), Insights
on leadership (pp. 170-184). New York: Wiley.
Kelly, G. A. (1955). The psychology of personal constructs (Vols 1 & 2). NY: Norton.
Kieffer, K. M. (1999). An introductory primer on the appropriate use of exploratory and
confirmatory factor analysis. Research in the Schools, 6, 75-92.
Kihlstrom, J. F., & Klein, S. B. (1994). The self as a knowledge structure. In R. S. Wyer
& T. K. Srull (Eds.), Handbook of Social Cognition (Vol. 1 (Basic Processes)).
New Jersey: Erlbaum.
Kirkpatrick, S. A., & Locke, E. A. (1996). Direct and indirect effects of three core
charismatic leadership components on performance and attitudes. Journal of
Applied Psychology, 81, 36-51.
Klein, S. B. (2001). A self to remember: A cognitive neuropsychological perspective on
how self creates memory and memory creates self. In C. Sedikides & M. B.
Brewer (Eds.), Individual self, relational self, collective self (pp. 25-46).
Philadelphia: Psychology Press/Taylor & Francis.
Kline, R. (1998). Principles and practice of structural equation modeling. New York:
Guildford Press.
Kolb, A. Y., & Kolb, D. A. (2005). Learning Styles and Learning Spaces: Enhancing
Experiential Learning in Higher Education. Academy of Management Learning &
Education, 4, 193.
Kolb, D. A. (1984). Experiential learning : experience as the source of learning and
development. Englewood Cliffs, N.J.: Prentice-Hall.
Koltko-Rivera, M. E. (2004). The Psychology of worldviews. Review of General
Psychology, 8(1), 3-58.
Kuhnert, K. W., & Lewis, P. (1987). Transactional and transformational leadership: A
constructive/developmental analysis. Academy of Management Review, 12, 648-
657.
LaPorte, T., R., & Consolini, P., M. (1991). Working in practice but not in theory:
Theoretical challenges of "high-reliability organizations". Journal of Public
Administration Research and Theory, 1(1), 19-48.
159
Lindzey, E., Holmes, V., & McCall, M. W. J. (1987). Key events in executive lives
(Techical Report No 32). Greensboro, N.C.: Center for Creative Leadership.
Locke, E. A., & Kirkpatrick, S. (1991). The essence of leadership : the four keys to
leading successfully. New York: Lexington Books.
London, M. (2002). Leadership development: Paths to self-insight and professional
growth. Mahwah, NJ: Elrbaum.
Lord, R. G., & Brown, D. J. (2001). Leadership, value, and subordinate self-concepts.
The Leadership Quarterly, 12, 133-152.
Lord, R. G., Foti, R. J., & De Vader, C. L. (1984). A test of leadership categorization
theory: Internal structure, information processing and leadership perceptions.
Organizational Behavior and Human Performance, 34, 343-378.
Lord, R. G., Foti, R. J., & Phillips, J. S. (1982). A theory of leadership categorization. In
J. G. Hunt, U. Sekaran & C. Schriesheim (Eds.), Leadership: Beyond
establishment views. Carbondale, IL: Southern Illinois Univerisity Press.
Lord, R. G., & Hall, R. J. (2005). Identity, deep structure and the development of
leadership skill. The Leadership Quarterly, 16(4), 591-615.
Luthans, F. (2002). Positive organizational behavior: Developing and managing
psychological strengths. Academy of Management Executive, 16, 57-72.
Luthans, F., Avey, J. B., Avolio, B. J., Norman, S., & Combs, G. (2006). Psychological
capital development: Towards a micro-intervention. Journal of Organizational
Behavior, 27, 387-393.
Luthans, F., Avolio, B., Avey, J., & Norman, S. (2007). Positive Psychological Capital:
Measurement and relationship with performance and satisfaction. Personnel
Psychology, 60, 541-572.
Luthans, F., & Avolio, B. J. (2003). Authentic leadership: A positive developmental
approach. In K. S. Cameron, J. E. Dutton & R. E. Quinn (Eds.), Positive
organizational scholarship (pp. 241-261). San Francisco: Barrett-Koehler.
Luthans, F., & Youssef, C. M. (2007). Emerging positive organizational behavior.
Journal of Management, 33, 321-349.
Luthans, F., Youssef, C. M., & Avolio, B. (2007). Psychological Capital. New York:
Oxford University Press.
Malinen, A. (2000). Towards the essence of adult experiential learning : a reading of the
theories of Knowles, Kolb, Mezirow, Revans and Schön. Jyväskylä, Finland:
University of Jyväskylä.
Markus, H. R., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954-
969.
Martin, L. L., & Tesser, A. (1996). Some ruminative thoughts. In R. S. Wyer (Ed.),
Ruminative thoughts (pp. 1-47). Mahwah, NJ: Lawrence Erlbaum.
Maurer, T., J., Weiss, E., M., & Barbeite, F., G. (2003). A model of involvement in work-
related learning and development activity: The effects of individual, situational,
motivational and age variables. Journal of Applied Psychology, 88(4), 707-724.
Meyer, D. E., & Schvandeveldt, R. W. (1971). Facilitation in recognizing pairs of words:
Evidence of a dependence between retrieval operations. Journal of Experimental
Psychology, 90(227-234).
Mezirow, J. (2000a). Learning as transformation : critical perspectives on a theory in
progress (1 ed.).
160
Mezirow, J. (2000b). Learning to think like an adult: Core concepts of transformation
theory. In J. Mezirow & Associates (Eds.), Learning as transformation: Critical
perspectives on a theory in progress (pp. 3-33). San Francisco: Jossey-Bass.
Mintzberg, H., Raisinghani, D., & Theoret, A. (1976). The structure of "unstructured"
decision processes. Administrative Science Quarterly, 21, 246-275.
Mischel, W. (1973). Toward a cognitive social reconceptualization of personality.
Psychological Review, 80, 253-283.
Mor, N., & Winquist, J. (2002). Self-focused attention and negative affect: A meta-
analysis. Psychological Bulletin, 128, 638-662.
Moskowitz, G. B. (2005). Social cognition : understanding self and others. New York:
Guilford Press.
Mulhauser, G. R. (1998). Mind out of matter : topics in the physical foundations of
consciousness and cognition. Dordrecht ; Boston: Kluwer Academic.
Mumford, M. D., & Connelly, M. S. (1991). Leaders as creators: Leader performance and
problem solving in ill-defined domains. Special Issue: Individual differences and
leadership: I. [STUDY TYPE: Theoretical]. The Leadership Quarterly, 2(4), 289-
315.
Mumford, M. D., Zaccaro, S. J., Harding, F. D., Jacobs, T. O., & Fleishman, E. A.
(2000). Leadership skills for a changing world: Solving complex social problems.
The Leadership Quarterly, 11(1), 11-36.
Munby, H., & Russell, T. (1990). Metaphors in the study of teachers' professional
knowledge. Theory into Practice, 29(2), 116-121.
Murphy, S. E., Blyth, D., & Fiedler, F. E. (1992). Cognitive resource theory and the
utilization of the leader's and group members' technical competence. [STUDY
TYPE: Empirical]. The Leadership Quarterly, 3(3), 237-255.
Muthen, B., & Muhen, L. K. (1998). MPLUS (Version 3.11). Los Angeles, California:
www.statmodel.com.
Neely, J. (1977). Semantic priming and retrieval from lexical memory: Roles of
inhibitionless spreading activation and limited capacity attention. Journal of
Experimental Psychology: General, 106(3), 226-254.
Newell, A., & Simon, H. A. (1972). Human problem-solving. Englewood Cliffs, N.J.:
Prentice-Hall.
Niedenthal, P. M., Halberstadt, J. B., & Setterlund, M. B. (1997). Being happy and seeing
"happy": Emotional state facilitates visual encoding. Cognition & Emotion, 11,
403-432.
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of
depressive episodes. Journal of Abnormal Psychology, 100, 569-582.
Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders and mixed
anxiety/depressive symptoms. Journal of Abnormal Psychology, 109, 504-511.
Nystedt, L., & Ljungbert, A. (2002). Facets of private and public self-consciousness:
Construct and discriminant validity. European Journal of Personality, 16(143-
159).
Offerman, L. R., Kennedy Jr., J. K., & Wirtz, P. W. (1994). Implicit leadership theories:
Content, structure, and generalizability. The Leadership Quarterly, 5(1), 43-58.
Olsen, C. L. (1974). Comparative Robustness of Six Tests in Multivariate Analysis of
Variance. Journal of the American Statistical 69(348), 894-908.
161
Pfeffer, J., & Davis-Blake, A. (1986). Administrative succession and organizational
performance: How administrative experience mediates the succession effect.
Academy of Management Journal, 29, 72-83.
Prentice-Dunn, S., & Rogers, R. W. (1989). Deindividuation and the self-regulation of
behavior. In P. B. Paulus (Ed.), The psychology of group influence (2nd ed) (pp.
87-109). Erlbaum, NJ: Hillsdale.
Pyszczynski, T., & Greenberg, J. (1987). Self-regulatory perseveration and the depressive
self-focusing style: A self-awareness theory of reactive depression. Psychological
Bulletin, 102, 122-138.
Quinn, R. E. (2005). Moments of greatness: Entering the fundamental state of leadership.
Harvard Business Review, 83(7), 74-84.
Reagans, R., & McEvily, B. (2003). Network Structure and Knowledge Transfer: The
Effects of Cohesion and Range. Administrative Science Quarterly, 48(2), 240.
Roberts, L. M., Dutton, J. E., Spreitzer, G. M., Heaphy, E. D., & Quinn, R. E. (2005).
Composing the reflected best-self portrait: building pathways for becoming
extraordinary in work organizations. Academy of Management Review, 30(4),
712-736.
Rousseau, D. M., & McCarthy, S. (2007). Educating Managers From an Evidence-Based
Perspective. Academy of Management Learning & Education, 6(1), 84-101.
Rubin, D. B. (1976). Inference and missing data. Biometrika, 61, 581-592.
Russell, A. L., & Cohen, L. M. (1997). The reflective colleague in email cyberspace: A
means for improving university instruction. Computers in Education, 29, 137-
145.
Schön, D. A. (1983). The reflective practitioner : how professionals think in action. New
York: Basic Books.
Shamir, B., & Eilam, G. (2005). "What's your story?" A life-stories approach to authentic
leadership development. The Leadership Quarterly, 16(3), 395-418.
Silvia, P. J. (2002). Self-Awareness and the Regulation of Emotional Intensity. Self &
Identity, 1(1), 3-10.
Silvia, P. J., & Duval, T. S. (2001). Objective Self-Awareness Theory: Recent Progress
and Enduring Problems. Personality & Social Psychology Review, 5(3), 230.
Smith, E. R. (1998). Mental representation and memory. In D. T. Gilbert, S. T. Fiske &
G. Lindzey (Eds.), Handbook of social psychology (4th ed) (Vol. 1, pp. 391-445).
Boston: McGraw-Hill.
Sosik, J. J. (2001). Self-other agreement on charismatic leadership: Relationships with
work attitudes and managerial performance. Group & Organizational
Management, 26, 484-511.
Sosik, J. J., & Dworakivsky, A. C. (1998). Self-concept based aspects of the charismatic
leader: More than meets the eye. [STUDY TYPE: Empirical]. The Leadership
Quarterly, 9(4), 503-526.
Sosik, J. J., Potosky, D., & Jung, D. I. (2002). Adaptive self-regulation: Meeting others'
expectation for leadership and performance. Journal of Social Psychology,
142(2), 211-232.
Stajkovic, A. D., & Luthans, F. (1998). Social cognitive theory and self-efficacy: Going
beyond traditional motivational and behavioral approaches. Organizational
Dynamics, 126, 62-74.
162
Sternberg, R. J. (2003). WICS: A Model of Leadership in Organizations. Academy of
Management Learning & Education, 2(4), 386.
Sternberg, R. J. (2004). Wisdom and giftedness. In L. Shavinina, V. & M. Ferrari (Eds.),
Beyond knowledge (pp. 169-186). Mahwah, New Jersey.: Lawrence Erlbaum
Associates.
Tay, C., Ang, S., & Dyne, L. V. (2006). Personality, Biographical Characteristics, and
Job Interview Success: A Longitudinal Study of the Mediating Effects of
Interviewing Self-Efficacy and the Moderating Effects of Internal Locus of
Causality. Journal of Applied Psychology, 91(2), 446-454.
Taylor, P., J., Russ-Eft, D., F., & Chan, D., W.L. (2005). A Meta-Analytic Review of
Behavior Modeling Training. Journal of Applied Psychology, 90(4), 692-709
Teasdale, J. D. (1999). Emotional processing, three modes of mind and the prevention of
relapse in depression. Behavior Research and Therapy, 37, S53-S77.
Teasdale, J. D., Segal, Z. V., Williams, J. M. G., Ridgeway, V., Soulsby, J., & Lau, M.
(2000). Prevention of relapse/recurrence in major depression by mindfulness-
based cognitive therapy. Journal of Consulting and Clinical Psychology, 68, 615-
623.
Trapnell, P., D., & Campbell, J., D. (1999). Private Self-Consciousness and the Five-
Factor Model of Personality: Distinguishing Rumination From Reflection.
Journal of Personality and Social Psychology, 76(2), 284-305.
Treynor, W., Gonzalez, R., & Nolen-Hoeksema, S. (2003). Rumination reconsidered: A
psychometric analysis. Cognitive Therapy and Research, 27, 247-259.
Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson
(Eds.), Organization of Memory (pp. 381-403). New York: Academic Press.
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases.
Science, 185, 1124-1131.
Velicer, W. F., Eaton, C. A., & Fava, J. L. (2000). Construct explication through factor or
component analysis: A review and evaluation of alternative procedures for
determining the number of factors or components. In R. D. Goffin & E. Helmes
(Eds.), Problems and solutions in human assessment: Honoring Douglas N.
Jackson at seventy. Norwell, MA: Kluwer Academic.
Watkins, E. (2004). Adaptive and maladaptive ruminative self-focus during emotional
processing. Behavior Research and Therapy, 42(9), 1037-1052.
Watkins, E., & Baracaia, S. (2002). Rumination and social problem-solving in
depression. Behavior Research and Therapy, 40, 1179-1189.
Watkins, E., & Moulds, M. (submitted). Adaptive and maladaptive ruminative self-focus
for problem solving in depression.
Wicklund, R. A. (1979). The influence of self-awareness on human behavior. American
Scientist, 67, 187-193.
Winograd, T. (1972). Understanding natural language. New York: Academic Press.
Wofford, J. C., & Goodwin, V. L. (1994). A cognitive interpretation of transactional and
transformational leadership theories. [STUDY TYPE: Theoretical Elaborate
model, 18 proposition]. The Leadership Quarterly, 5(2), 161-186.
Wofford, J. C., Goodwin, V. L., & Whittington, J. L. (1998). A field study of a cognitive
approach to understanding transformational and transactional leadership.
[STUDY TYPE: Empirical]. The Leadership Quarterly, 9(1), 55-84.
163
Zweig, D., & Webster, J. (2004). What are we measuring? An examination of the
relationships between goal orientation, the big-five personality traits and
performance intentions. Personality and Individual Differences, 36(7), 1693-
1708.
Zwick, W. R., & Velicer, W. F. (1986). Factors influencing five rules for determining the
number of components to retain. Psychological Bulletin, 99, 432-442.